Group Research Projects

Discover how differential equations power your engineering specialty. Pick a topic, form a team, research it, and teach the class what you learned.

Project Objective

The goal of this project is to connect what you learn in this DE course to real-world problems in your own engineering specialty. By researching a topic where differential equations play a central role, you will see first-hand that DEs are not abstract math — they are the language that engineers use to describe, predict, and design the systems around us.

مَنْ سَلَكَ طَرِيقًا يَلْتَمِسُ فِيهِ عِلْمًا سَهَّلَ اللَّهُ لَهُ بِهِ طَرِيقًا إِلَى الْجَنَّةِ
"Whoever travels a path in search of knowledge, Allah will make easy for him a path to Paradise."
— Prophet Muhammad ﷺ (Sahih Muslim)

How It Works

Form a group of three students (ideally from the same or related specialty). Choose one project idea from the list below — or propose your own topic with instructor approval. Research the topic together, and prepare a 10–15 minute presentation to share what you learned with the rest of the class.

👥

Team Size

Groups of 3 students

🎯

Choose a Topic

Pick from the ideas below or propose your own

🔬

Research

Understand the DE model, solve or simulate it, explain results

🎤

Present

10–15 minute presentation to the class

What Your Presentation Should Include

Every presentation, regardless of the specialty, should address the following:

  1. The Real-World Problem — What physical, engineering, or scientific problem are you studying? Why does it matter?
  2. The DE Model — What differential equation describes the system? Explain each term and variable.
  3. Solution or Simulation — Solve the DE analytically (if possible) or show a numerical/graphical solution. Interpret what the solution tells you.
  4. Connection to Your Major — How do engineers in your field use this model? What design decisions does it inform?
  5. What You Learned — Key takeaways and anything that surprised you.

Grading Rubric

CriterionWeightDescription
DE Understanding30%Correctly state the DE, explain terms, show solution or simulation
Real-World Connection25%Clear link between the math and the engineering application
Presentation Quality20%Organized slides, clear speaking, good visuals, within time limit
Depth of Research15%Goes beyond surface-level; cites sources; shows genuine understanding
Team Collaboration10%All members contribute meaningfully during the presentation
📋

Step 1: Select Your Section

Choose your lecture section below. This ensures no two groups in the same section pick the same project.

Step 2: Choose Your Specialty

Click your engineering department to jump directly to its project ideas.

Electrical Engineering

— 10 project ideas
EE-01

RLC Circuit Transient Response

Ch 2 Introductory Conceptual
+
Background

RLC circuits are fundamental to electrical engineering, found in power supplies, filters, and radio tuners. When a switch closes or opens, the circuit undergoes a transient response before reaching steady state. Understanding this transient behavior is critical for designing circuits that respond appropriately to sudden changes.

The DE: \(L\frac{d^2i}{dt^2} + R\frac{di}{dt} + \frac{i}{C} = V(t)\) where \(L\) is inductance, \(R\) is resistance, \(C\) is capacitance, and \(V(t)\) is the applied voltage.

What to Investigate
  • How does the damping ratio determine whether the circuit oscillates or settles smoothly to steady state?
  • What is resonance, and why does excessive current flow at the natural frequency?
  • How do engineers choose component values to achieve overdamped, critically damped, or underdamped responses?
  • Why are RLC circuits used in radio receivers to select specific frequencies?

Presentation tip: Create a visual showing how current vs. time graphs change as you vary the damping ratio. Show oscillatory vs. non-oscillatory cases side-by-side.

EE-02

RC Circuit and Time Constants

Ch 1 Introductory Conceptual
+
Background

RC circuits govern the charging and discharging of capacitors, appearing in camera flash circuits, touch screens, and analog filters. The time constant determines how quickly a capacitor responds to input changes. Smartphones use RC circuits to filter finger touches and distinguish them from noise.

The DE: \(RC\frac{dV_C}{dt} + V_C = V_{in}(t)\) where \(V_C\) is the capacitor voltage and \(V_{in}\) is the input voltage.

What to Investigate
  • What is the time constant \(\tau = RC\), and why does it determine charging/discharging speed?
  • Why does the capacitor voltage never instantaneously match the input?
  • How can touchscreen engineers use the time constant to filter out noise while preserving real touches?
  • What happens if you apply an AC signal to an RC circuit instead of a step input?

Presentation tip: Show actual smartphone touch calibration data or camera flash timing, then explain how the time constant ensures proper operation.

EE-03

Transfer Functions and Laplace in Circuit Analysis

Ch 3 Moderate Analytical
+
Background

Transfer functions and Laplace transforms are essential tools for circuit design and analysis. They allow engineers to work in the frequency domain (s-domain) instead of the time domain, making complex circuits much easier to analyze. Control systems, audio equalizers, and antenna filters all rely on transfer function concepts.

The Concept: A transfer function \(H(s) = \frac{V_{out}(s)}{V_{in}(s)}\) describes how a circuit responds to different frequencies, with poles and zeros determining stability and resonance behavior.

What to Investigate
  • What do poles and zeros of a transfer function tell you about circuit stability and frequency response?
  • Why does working in the s-domain transform differential equations into simpler algebraic equations?
  • How do engineers use Bode plots to visualize frequency response without solving equations?
  • What does it mean when poles cross the right half-plane, and why is this dangerous?
EE-04

DC Motor Speed Control

Ch 4 Moderate Conceptual + Simulation
+
Background

DC motors are found everywhere—from power tools to electric vehicles. Motor speed depends on applied voltage and load torque. The relationship between electrical input (voltage) and mechanical output (shaft speed) forms a system of coupled differential equations that describe the electromechanical interaction.

The DE System: Electrical: \(L\frac{di}{dt} + Ri + K_b\omega = V(t)\); Mechanical: \(J\frac{d\omega}{dt} + b\omega = K_t i - T_L\) where \(\omega\) is speed, \(K_b\) is back-EMF constant, and \(K_t\) is torque constant.

What to Investigate
  • Why does the back-EMF (counter-voltage) increase as the motor speeds up, eventually limiting current?
  • How do the electrical and mechanical time constants differ, and which dominates motor response?
  • What role does inertia play in how quickly a motor can accelerate or decelerate?
  • How would you design a control system to maintain constant speed despite changing load?

Presentation tip: Simulate or demonstrate a motor starting with and without load, showing how the speed curve changes with different applied voltages.

EE-05

Transmission Line Equations — Telegrapher's Equations

Ch 4 Advanced Analytical
+
Background

Power transmission lines, coaxial cables, and telephone wires are governed by coupled partial differential equations called the Telegrapher's Equations. These equations describe how voltage and current propagate down a cable as waves. Understanding transmission lines is critical for power system design and high-speed digital communication.

The PDE: \(\frac{\partial V}{\partial z} = -L\frac{\partial i}{\partial t} - Ri\) and \(\frac{\partial i}{\partial z} = -C\frac{\partial V}{\partial t} - GV\) where \(L, C, R, G\) are per-unit-length parameters.

What to Investigate
  • How do voltage and current waves propagate down a transmission line, and at what velocity?
  • What is characteristic impedance, and why must it be matched to prevent reflections?
  • What happens when a signal encounters the end of a cable (open or short circuit)?
  • Why do high-speed digital signals require careful cable management and impedance matching?

Presentation tip: Animate voltage and current waves traveling down a transmission line, showing what happens at the termination.

EE-06

Phase-Locked Loops in Communication Systems

Ch 4 Advanced Conceptual + Simulation
+
Background

Phase-locked loops (PLLs) synchronize oscillators to incoming signals and are essential components in radio receivers, GPS systems, and clock recovery circuits. A PLL continuously adjusts a local oscillator's frequency to match an incoming signal's frequency, creating a feedback control system governed by nonlinear differential equations.

The DE: \(\frac{d\theta_{vco}}{dt} = \omega_0 + K_p e(t) + K_i \int e(\tau)d\tau\) where \(e(t)\) is the phase error and \(K_p, K_i\) are proportional and integral gains.

What to Investigate
  • How does negative feedback in a PLL help synchronize the local oscillator to an incoming signal?
  • What determines the lock range—the range of input frequencies the PLL can synchronize to?
  • Why is the transient response of a PLL (lock-in time) important in wireless receivers?
  • What advantages do proportional-integral (PI) controllers provide over simple proportional control?
EE-07

Power Grid Stability and Swing Equation

Ch 2 Advanced Conceptual + Simulation
+
Background

Power grids operate at a constant frequency (60 Hz in North America, 50 Hz elsewhere), and generators must maintain synchronism despite varying loads and disturbances. The swing equation governs generator rotor dynamics and is central to power system stability analysis. A blackout can result from instability in this nonlinear system.

The DE: \(2H\frac{d^2\delta}{dt^2} + D\frac{d\delta}{dt} = P_m - P_e(\delta)\) where \(\delta\) is rotor angle, \(H\) is inertia constant, \(P_m\) is mechanical power, and \(P_e\) is electrical power.

What to Investigate
  • What role does generator inertia play in the power grid's ability to withstand sudden disturbances?
  • Why is synchronism stability a nonlinear problem, and what happens when generators lose synchronism?
  • How do protective relays detect instability and isolate failing components before cascading failure?
  • What is the critical clearing time, and why is it important for grid reliability?

Presentation tip: Simulate a sudden fault (like a short circuit) and show how different inertia levels affect grid stability.

EE-08

Semiconductor Physics — Drift-Diffusion Equation

Ch 5 Advanced Analytical
+
Background

Transistors, diodes, and solar cells operate based on charge carrier movement through semiconductor materials. The drift-diffusion equation describes how electrons and holes move due to both electric fields (drift) and concentration gradients (diffusion). This partial differential equation is the foundation of semiconductor device modeling and design.

The PDE: \(\frac{\partial n}{\partial t} = \frac{1}{q}\frac{\partial J_n}{\partial x} - R\) where \(J_n = qD_n\frac{\partial n}{\partial x} - q\mu_n n E\) is the electron current density.

What to Investigate
  • How do drift (electric field) and diffusion (concentration gradient) compete in determining carrier motion?
  • What is the Einstein relation, and why does it connect diffusivity and mobility?
  • How do pn-junctions create the depletion region that makes diodes and transistors work?
  • Why do temperature and doping concentration profoundly affect semiconductor device performance?
EE-09

EM Wave Propagation — Maxwell to Wave Equation

Ch 4 Advanced Analytical
+
Background

Electromagnetic waves—radio, microwaves, light—are the foundation of wireless communication, radar, and fiber optics. Maxwell's equations describe how electric and magnetic fields couple and propagate. Combining these equations yields the wave equation, revealing that EM waves travel at the speed of light and their wavelength and frequency are related by \(c = \lambda f\).

The Wave Equation: \(\nabla^2 E - \mu_0 \epsilon_0 \frac{\partial^2 E}{\partial t^2} = 0\) with wave speed \(v = \frac{1}{\sqrt{\mu_0 \epsilon_0}} = c\).

What to Investigate
  • How do changing electric and magnetic fields continuously generate each other, sustaining wave propagation?
  • Why do radio waves at different frequencies propagate differently through materials?
  • What is impedance matching, and why do antennas have specific shapes and sizes?
  • How does the wave equation explain attenuation and reflection in conductors versus insulators?

Presentation tip: Show how antenna design relates to wavelength, explaining why a smartphone's 4G antenna is much shorter than a radio station's FM antenna.

EE-10

Signal Processing and Convolution — Impulse Response

Ch 3 Moderate Conceptual + Simulation
+
Background

Linear time-invariant (LTI) systems, such as audio filters, equalizers, and communication channels, can be completely characterized by a single function: the impulse response. Convolution combines the input signal with the impulse response to predict any output. This principle is fundamental to audio processing, image filtering, and signal denoising.

The Concept: \(y(t) = \int_{-\infty}^{\infty} h(\tau) u(t - \tau) d\tau\) where \(h(t)\) is the impulse response and \(u(t)\) is the input signal.

What to Investigate
  • What does the impulse response tell you about how a system will respond to any input signal?
  • Why does convolution mathematically describe how systems filter and shape signals?
  • How can the Laplace transform turn convolution (an integral) into simple multiplication?
  • How do audio engineers use impulse responses to simulate room acoustics or famous recording studios?

Presentation tip: Record a room's impulse response with a speaker and microphone, then show how convolving it with music creates a realistic reverberation effect.

⚙️

Mechanical Engineering

— 10 project ideas
ME-01

Mass-Spring-Damper Vibration Analysis

Ch 2 Introductory Conceptual
+
Background

The mass-spring-damper system is the foundation of vibration analysis and appears everywhere: car suspensions, earthquake-resistant buildings, and vibration isolation systems. When a mass is displaced from equilibrium, springs restore it and dampers dissipate energy. Understanding the balance between these forces explains everything from smooth rides to structural failures.

The DE: \(m\frac{d^2x}{dt^2} + c\frac{dx}{dt} + kx = F(t)\) where \(m\) is mass, \(c\) is damping coefficient, \(k\) is spring stiffness, and \(F(t)\) is applied force.

What to Investigate
  • How does the damping ratio determine whether vibrations oscillate or decay monotonically?
  • What is resonance, and why can even small periodic forces cause large oscillations near the natural frequency?
  • Why do car suspensions need both springs and shock absorbers—what does each provide?
  • How would you design a suspension to isolate a sensitive instrument from floor vibrations?

Presentation tip: Show videos of car suspensions responding to bumps, then explain the physics using your differential equation model.

ME-02

Newton's Law of Cooling

Ch 1 Introductory Conceptual
+
Background

Newton's Law of Cooling describes how objects exchange heat with their surroundings. It governs heating, cooling, and thermal comfort in buildings, refrigeration systems, and even forensic science. When your coffee cools down, when an oven heats food, or when engineers design HVAC systems, Newton's law is at work.

The DE: \(\frac{dT}{dt} = -h(T - T_{ambient})\) where \(T\) is object temperature, \(T_{ambient}\) is surroundings temperature, and \(h\) is a heat transfer coefficient.

What to Investigate
  • Why is the cooling rate proportional to the temperature difference, not the absolute temperature?
  • What is the time constant for cooling, and how does it depend on material properties and environment?
  • How do thermally insulated containers slow cooling, and what does this reveal about the heat transfer coefficient?
  • How can forensic scientists use Newton's law to estimate time of death from body temperature?

Presentation tip: Measure the temperature of hot water or coffee over time and fit your data to the exponential solution.

ME-03

Projectile Motion with Air Resistance

Ch 1 Moderate Conceptual + Simulation
+
Background

Projectile motion without air resistance is idealized high school physics, but real objects experience drag. Whether it's baseball trajectory in sports, artillery range calculations, or skydivers reaching terminal velocity, air resistance fundamentally changes how objects move. Understanding drag is essential for sports engineering, aerospace design, and safety systems.

The DEs: \(m\frac{dv_x}{dt} = -bv_x\) and \(m\frac{dv_y}{dt} = -mg - bv_y\) where \(b\) is the drag coefficient and \(g\) is gravitational acceleration.

What to Investigate
  • How does air resistance change the optimal launch angle for maximum range?
  • Why do objects reach terminal velocity, and what determines it?
  • How do baseball pitches and curveballs exploit aerodynamic effects?
  • Why is skydiver acceleration initially high but eventually reaches near-zero?

Presentation tip: Compare trajectories with and without drag for a baseball or basketball, explaining why real sports don't follow parabolic paths.

ME-04

Pendulum Dynamics — Linear to Chaotic

Ch 2 Advanced Conceptual + Simulation
+
Background

The simple pendulum is deceptively complex. For small angles, the equation linearizes and behaves predictably. But for larger angles or with driving forces, the pendulum exhibits chaos—sensitive dependence on initial conditions where tiny differences lead to completely different outcomes. This nonlinear phenomenon illustrates why weather and complex systems are unpredictable.

The Nonlinear DE: \(\frac{d^2\theta}{dt^2} + \frac{g}{L}\sin\theta = 0\) (exact); approximation: \(\frac{d^2\theta}{dt^2} + \frac{g}{L}\theta = 0\) (small angle).

What to Investigate
  • Why is the small-angle approximation valid for small angles but fails for large angles?
  • How does nonlinearity prevent superposition—can you add two solutions to get another solution?
  • What is chaos, and how does a driven, damped pendulum transition from regular to chaotic motion?
  • Why are Lyapunov exponents important for understanding long-term predictability?

Presentation tip: Simulate pendulum trajectories with slightly different initial conditions, showing how they diverge exponentially in chaotic regimes.

ME-05

Heat Conduction — Fourier's Law and Heat Equation

Ch 4 Moderate Analytical
+
Background

Heat conduction governs temperature distribution within solid materials. From cooking food to designing semiconductor cooling systems, understanding how heat diffuses through materials is critical. The heat equation is a fundamental PDE in engineering and shows how temperature gradients drive heat flow and eventually homogenize temperature.

The PDE: \(\rho c\frac{\partial T}{\partial t} = k\nabla^2 T + q\) where \(\rho\) is density, \(c\) is specific heat, \(k\) is thermal conductivity, and \(q\) is internal heat generation.

What to Investigate
  • How does thermal conductivity determine how fast temperature changes propagate through a material?
  • Why does the heat equation smooth out temperature discontinuities over time?
  • How do engineers design heat sinks for electronics to keep temperatures manageable?
  • What is the Fourier number, and why does it determine whether transient or steady-state analysis applies?
ME-06

Fluid Flow in Pipes — Navier-Stokes Connection

Ch 2 Advanced Analytical + Simulation
+
Background

Fluid flow through pipes is central to engineering: water distribution, blood circulation, oil pipelines, and hydraulic systems all rely on understanding how viscosity, pressure, and flow rate interact. The Navier-Stokes equations govern fluid motion, and for fully developed pipe flow, they yield the elegant Hagen-Poiseuille formula predicting laminar flow profiles.

The Result (Hagen-Poiseuille): \(Q = \frac{\pi R^4 \Delta P}{8\mu L}\) where \(Q\) is flow rate, \(\Delta P\) is pressure drop, \(\mu\) is viscosity, and \(R, L\) are pipe radius and length.

What to Investigate
  • Why does flow rate depend on the fourth power of pipe radius—what does this reveal about fluid resistance?
  • How does viscosity resistance dominate in laminar flow but become less important in turbulent flow?
  • What is the Reynolds number, and how does it predict whether flow will be laminar or turbulent?
  • How do doctors understand blood flow and design better stents for clogged arteries?

Presentation tip: Show how blood flow rates change in stenosed (partially blocked) arteries and why the r^4 dependence makes small blockages dangerous.

ME-07

Shaft Torsion and Torsional Vibration

Ch 2 Moderate Conceptual + Simulation
+
Background

Rotating shafts in turbines, pumps, and motor drivetrains experience torsional vibrations—twisting oscillations that can cause fatigue failure. Torsional stiffness and rotational inertia determine natural frequencies. Resonance in torsional vibration has caused catastrophic failures in turbines and submarines, making this analysis critical for safe design.

The DE: \(I_p\frac{d^2\theta}{dt^2} + c\frac{d\theta}{dt} + k_t\theta = T(t)\) where \(I_p\) is polar moment of inertia, \(k_t\) is torsional stiffness, and \(T(t)\) is applied torque.

What to Investigate
  • How do torsional rigidity and rotational inertia determine the natural frequency of torsional vibration?
  • Why can periodic torque inputs (like from reciprocating engines) excite torsional resonance?
  • What is a torsional damper, and how does it prevent dangerous resonance?
  • How do engineers avoid operating turbines and compressors near torsional natural frequencies?
ME-08

Rocket Propulsion — Tsiolkovsky Equation

Ch 1 Moderate Conceptual
+
Background

Rockets propel spacecraft by ejecting mass at high velocity. Unlike cars that push against the road, rockets carry all their propellant and must expel it to accelerate. The Tsiolkovsky rocket equation relates velocity change to exhaust velocity and fuel mass. It governs whether missions reach orbit, escape Earth's gravity, or achieve deep space goals.

The Tsiolkovsky Equation: \(\Delta v = v_e \ln\left(\frac{m_0}{m_f}\right)\) where \(v_e\) is exhaust velocity, \(m_0\) is initial mass (with fuel), and \(m_f\) is final mass (empty).

What to Investigate
  • Why does velocity change depend logarithmically on mass ratio, not linearly?
  • What is the ideal velocity increment (delta-v) needed to reach lunar orbit, escape velocity, Mars transfer?
  • Why do multi-stage rockets provide better performance than single-stage designs?
  • How do gravity losses and atmospheric drag reduce actual achieved velocity below theoretical delta-v?

Presentation tip: Calculate delta-v requirements for various space missions and explain why Saturn V needed so much fuel to land humans on the Moon.

ME-09

Lumped Capacitance — Transient Heat Transfer

Ch 1 Moderate Conceptual
+
Background

When a solid object is suddenly exposed to different temperature (quenching steel, cooling electronics), temperature first drops rapidly at the surface, then slowly throughout the interior. The lumped capacitance method assumes the entire object remains at uniform temperature, valid when internal conduction is fast compared to surface cooling. This simplifies transient heat transfer analysis significantly.

The DE: \(\rho V c\frac{dT}{dt} = hA(T_{ambient} - T)\) where \(\rho V\) is total mass, \(c\) is specific heat, \(hA\) is total heat transfer conductance.

What to Investigate
  • What is the Biot number, and when is lumped capacitance valid versus needing spatial analysis?
  • Why does a small object cool faster than a large one with the same material properties?
  • How do metallurgists use quenching to harden steel, and what differential equations guide the process?
  • How does surface heat transfer coefficient (coating, fluid type) dominate cooling rate?

Presentation tip: Record temperature at the surface and center of a large object during cooling, comparing to lumped capacitance predictions.

ME-10

PID Control of Mechanical Systems

Ch 4 Advanced Conceptual + Simulation
+
Background

PID (Proportional-Integral-Derivative) control is ubiquitous in engineering: cruise control in cars, temperature regulation in ovens, and robotic arm positioning. A PID controller continuously compares actual output to desired setpoint and adjusts the control input. This feedback system governs transient response, steady-state error, and stability—critical for safe, responsive systems.

PID Law: \(u(t) = K_p e(t) + K_i \int e(\tau)d\tau + K_d \frac{de}{dt}\) where \(e(t) = \text{setpoint} - \text{output}\) is the error.

What to Investigate
  • What does each term (P, I, D) do: proportional responds to current error, integral corrects steady-state offset, derivative anticipates change?
  • How do control gains (Kp, Ki, Kd) affect stability, settling time, and overshoot?
  • Why is tuning a PID controller an art—what design rules (Ziegler-Nichols) help?
  • How do engineers prevent integrator windup when the system saturates?

Presentation tip: Simulate cruise control—show how different PID gains produce oscillations, sluggish response, or smooth tracking.

🏗️

Civil Engineering

— 8 project ideas
CE-01

Beam Deflection Under Load

Ch 2: 2nd-Order Moderate Conceptual + Simulation
+
Background

The Euler-Bernoulli beam equation is the foundation of structural mechanics and governs how beams deflect under applied loads. Every bridge, building, and floor system relies on predictions from this differential equation to ensure safety and performance. Understanding beam deflection is essential for designing structures that can support loads without excessive displacement or failure.

The DE: \(\frac{d^2}{dx^2}\left(EI\frac{d^2y}{dx^2}\right) = w(x)\), where \(EI\) is flexural rigidity, \(y\) is deflection, and \(w(x)\) is distributed load.

What to Investigate
  • How boundary conditions (fixed, pinned, cantilever) change the deflection pattern and determine structural safety factors
  • Why material properties (\(E\) = Young's modulus, \(I\) = moment of inertia) control stiffness and how engineers choose materials for different applications
  • The relationship between load distribution and maximum deflection, and why certain load patterns are more critical than others
  • How this equation guides real-world decisions in bridge span design, skyscraper heights, and floor vibration limits

Presentation tip: Use numerical simulations or interactive visualizations to show how changing material properties or load types affects the deflection curve in real time. Compare predictions with actual measurements from engineered structures.

CE-02

Groundwater Flow — Darcy's Law

Ch 1: 1st-Order Introductory Conceptual
+
Background

Darcy's Law describes how water flows through soil and rock layers, making it fundamental to groundwater engineering and environmental protection. The rate at which water moves through the subsurface determines where contaminants spread, how quickly wells can be drained, and whether landfills will contaminate drinking water supplies. This simple yet powerful principle underlies decisions affecting water safety for millions of people.

The DE: \(\frac{dh}{dt} = -K\frac{d^2h}{dx^2}\), where \(h\) is hydraulic head, \(K\) is permeability, and the equation describes diffusion of pressure through saturated soil.

What to Investigate
  • How soil permeability varies with grain size and how engineers measure or estimate \(K\) for different soil types
  • Contaminant transport: why pollutants follow the same flow paths as water and how DE models predict spread over decades
  • Design of wells and pumping systems: how hydraulic head gradients determine extraction rates and radius of influence
  • Real-world application to landfill design, superfund site remediation, and geothermal systems

Presentation tip: Create a visual simulation showing how a contaminant plume spreads through soil layers over time, connecting the differential equation directly to environmental risk assessment and decision-making.

CE-03

Earthquake Response of Buildings — Seismic Engineering

Ch 2: 2nd-Order Advanced Conceptual + Simulation
+
Background

Seismic response analysis models how buildings vibrate and respond to earthquake ground motion using forced vibration equations. The natural frequency and damping of a structure determine whether it amplifies or attenuates seismic waves, making the difference between minor damage and catastrophic failure. Modern seismic design uses these differential equations to predict resonance and design damping systems that protect lives and infrastructure.

The DE: \(m\ddot{x} + c\dot{x} + kx = -m\ddot{x}_g(t)\), where \(m\) is mass, \(c\) is damping, \(k\) is stiffness, and \(\ddot{x}_g(t)\) is ground acceleration from the earthquake.

What to Investigate
  • Resonance and natural frequency: why buildings have critical frequencies at which they vibrate most intensely and how this relates to earthquake characteristics
  • Damping mechanisms: how dampers, base isolators, and tuned mass dampers reduce seismic response and enable tall buildings in earthquake zones
  • Multi-story coupling: how forces propagate through building floors and why different stories experience different accelerations
  • Design standards and codes: how seismic DE analysis informs building codes and protection strategies in high-risk regions

Presentation tip: Use earthquake recordings and animations to show actual ground motion input, then demonstrate how different building designs (with varying frequencies and damping) respond to the same earthquake. Contrast historical failures with modern designs.

CE-04

Settlement of Foundations — Soil Consolidation Theory

Ch 1: 1st-Order Moderate Conceptual
+
Background

Terzaghi's consolidation equation describes how soil compresses over time as water drains from pores under the weight of a building. Buildings constructed on soft clay can settle unevenly over months or years, causing cracks, structural damage, and alignment problems in adjacent structures. Engineers use this differential equation to predict settlement timing and magnitude, guiding decisions about pile depths, raft foundations, and preloading strategies.

The DE: \(\frac{\partial u}{\partial t} = c_v\frac{\partial^2 u}{\partial z^2}\), where \(u\) is excess pore pressure, \(c_v\) is coefficient of consolidation, and \(z\) is depth in the soil layer.

What to Investigate
  • Time-dependent nature of settlement: why buildings don't settle all at once and how long the consolidation process actually takes
  • Soil properties that control consolidation rate, including permeability and compressibility characteristics of different clay types
  • Primary vs. secondary consolidation: why differential settlement between adjacent buildings can cause serious problems
  • Practical design responses: pile foundations, raft footings, and preloading techniques used to mitigate settlement in real projects

Presentation tip: Show case studies of buildings that experienced unexpected settlement (with photographs and timelines), then demonstrate how Terzaghi's equation could have predicted the settlement behavior.

CE-05

Flood Routing — How Water Moves Through River Systems

Ch 1: 1st-Order Moderate Conceptual + Simulation
+
Background

Flood routing equations (simplified Saint-Venant equations) model how a flood wave propagates downstream through a river system over hours or days. Accurate flood prediction allows communities to evacuate, activate flood barriers, and manage dam releases strategically to minimize damage. These differential equations form the basis of flood forecasting systems that protect lives and enable emergency response.

The DE: \(\frac{\partial Q}{\partial x} + \frac{\partial A}{\partial t} = 0\) (continuity), combined with momentum equations governing how discharge \(Q\) and cross-sectional area \(A\) change with position and time.

What to Investigate
  • Wave speed vs. water speed: why flood waves travel faster than the average water velocity in the river
  • Channel geometry effects: how river width, depth, and roughness influence flood wave shape and arrival time at downstream locations
  • Reservoir routing and dam operations: how engineers release water strategically to flatten the flood peak and reduce downstream damage
  • Real-world forecasting systems: how flood prediction models integrate rainfall data, river measurements, and DE solutions for emergency management
CE-06

Traffic Flow Theory — The Lighthill-Whitham-Richards Model

Ch 1: 1st-Order Advanced Conceptual
+
Background

The Lighthill-Whitham-Richards (LWR) model treats traffic as a compressible fluid, using conservation laws and partial differential equations to predict how congestion waves form and propagate. Traffic engineers use this model to understand phantom jams (congestion with no apparent cause), optimize signal timing, and design highway capacity. This perspective reveals why simple interventions like staggered signal timing or ramp metering can smooth flow across entire cities.

The DE: \(\frac{\partial \rho}{\partial t} + \frac{\partial (\rho v)}{\partial x} = 0\), where \(\rho\) is vehicle density and \(v = v(\rho)\) is the speed-density relationship that captures how drivers slow down when traffic is dense.

What to Investigate
  • Fundamental diagram: the relationship between density, flow, and speed, and why traffic becomes unstable beyond a critical density
  • Shock waves in traffic: how congestion waves propagate backward against traffic flow, even when all drivers are moving forward
  • Metastability and phantom jams: why small perturbations (a sudden lane change, a merge) can trigger large-scale congestion that persists long after the initiating event
  • Control strategies: how ramp metering, variable speed limits, and signal coordination exploit the differential equation structure to suppress congestion waves

Presentation tip: Use video simulations or real traffic data to show congestion waves forming and propagating, then demonstrate how the LWR model predicts their behavior and how simple interventions suppress them.

CE-07

Concrete Curing and Strength Gain — Maturity Method

Ch 1: 1st-Order Introductory Conceptual
+
Background

The maturity method uses an Arrhenius-based differential equation to predict concrete strength development as a function of temperature and time. Construction schedules depend on knowing when concrete is strong enough to remove formwork or support loads, and the maturity method allows this to be predicted without waiting for standard 28-day tests. This enables faster construction and earlier project advancement on time-critical work.

The DE: \(\frac{d\alpha}{dt} = Ae^{-E_a/RT(t)}\), where \(\alpha\) is the degree of hydration, \(E_a\) is activation energy, \(R\) is the gas constant, and \(T(t)\) is temperature as a function of time.

What to Investigate
  • Temperature dependence of curing: why cold concrete cures slowly and how ambient conditions affect construction schedules
  • Strength-maturity relationships: how engineers correlate the cumulative thermal effect to actual compressive strength for decision-making
  • Early-age cracking: how rapid strength gain at high temperatures can prevent thermal cracking, and conversely how slow gain in cold weather creates risk
  • Practical construction applications: accelerated curing, steam curing, and winter construction strategies guided by maturity equations
CE-08

Cable Dynamics in Suspension Bridges — The Catenary Equation

Ch 2: 2nd-Order Advanced Conceptual + Simulation
+
Background

The catenary equation describes the shape of a cable hanging under its own weight and is central to understanding suspension bridge design. While many assume cables hang in parabolic curves, they actually form catenaries — a different curve entirely. Understanding this distinction is essential for computing cable tensions, predicting sag, and designing anchorages that safely support the massive live loads of traffic.

The DE: \(\frac{d^2y}{dx^2} = \frac{w}{H}\sqrt{1 + \left(\frac{dy}{dx}\right)^2}\), where \(w\) is weight per unit length, \(H\) is horizontal tension, and \(y\) is the cable shape.

What to Investigate
  • Catenary vs. parabola: why suspended cables form catenaries rather than parabolic arcs, and the physical meaning of this distinction
  • Cable tension distribution: how tension varies along the cable span and why anchorages must handle enormous forces at both ends
  • Dynamic response: how wind, earthquakes, and traffic loads cause cables to vibrate and why understanding cable dynamics is critical for bridge safety
  • Design implications: how catenary analysis guides choices in cable diameter, sag-to-span ratio, and anchorage design in real suspension bridges

Presentation tip: Create interactive visualizations showing how changing cable weight, span, or tension alters the catenary shape. Compare theoretical curves with photographs of actual suspension bridges and discuss why the Golden Gate Bridge looks like a catenary, not a parabola.

💻

Computer Engineering

— 8 project ideas
CS-01

Machine Learning and Gradient Descent

Ch 1: 1st-Order Advanced Conceptual
+
Background

Gradient descent, the optimization algorithm behind all neural networks, can be viewed as a continuous-time first-order ODE when learning rates are infinitesimal. This perspective reveals why gradient descent converges (or diverges), how momentum methods work, and connections to classical dynamical systems theory. Understanding this deep link between DEs and machine learning is transforming how researchers design better learning algorithms.

The DE: \(\frac{d\mathbf{w}}{dt} = -\nabla L(\mathbf{w})\), where \(\mathbf{w}\) are network weights and \(L(\mathbf{w})\) is the loss function; discrete gradient descent is a numerical integration of this ODE.

What to Investigate
  • Convergence analysis: why the ODE perspective explains convergence rates and stability properties of different optimizers
  • Momentum methods: how Nesterov acceleration and Adam optimization correspond to modified ODEs with different damping and acceleration terms
  • Loss landscape geometry: how the structure of \(\nabla L\) (critical points, saddle points) determines optimizer behavior and generalization
  • Modern research frontiers: Neural ODEs, which use differential equations as neural network layers, and implications for physics-informed machine learning

Presentation tip: Visualize the loss landscape as a surface in 2D or 3D, show how different optimizers trace different trajectories through it, and demonstrate how faster convergence relates to stability properties of the underlying ODE.

CS-02

Network Traffic and Queueing Theory

Ch 1: 1st-Order Moderate Conceptual + Simulation
+
Background

Queueing theory uses systems of first-order ODEs (birth-death processes) to model how packets accumulate at routers, jobs wait for processor cycles, and users queue for resources. These differential equations predict latency, throughput, and stability conditions for networks under varying load. Cloud providers and content delivery networks rely on queueing models to provision servers, set capacity limits, and prevent congestion collapses.

The DE: \(\frac{dP_n}{dt} = \lambda P_{n-1} - (\lambda + \mu)P_n + \mu P_{n+1}\), where \(P_n(t)\) is the probability of \(n\) items in queue, \(\lambda\) is arrival rate, and \(\mu\) is service rate.

What to Investigate
  • Stability conditions: when does a queue become unbounded (instability) versus reaching steady-state, and what does this mean for network design
  • Little's Law and performance metrics: how average queue length, waiting time, and throughput relate to arrival and service rates
  • Server load balancing: how distributing requests across multiple servers using queueing models improves response time and prevents bottlenecks
  • Real-world application to data center provisioning, CDN design, and congestion control in TCP networks

Presentation tip: Simulate a network router with adjustable arrival rate and number of servers, showing how queue length and latency evolve according to the ODE, and demonstrating the effect of overload conditions.

CS-03

Digital Signal Processing — From Analog Filters to Digital

Ch 2: 2nd-Order Advanced Conceptual
+
Background

Analog filters are governed by differential equations, while digital filters are designed by discretizing these same equations into difference equations. Understanding this relationship is fundamental to signal processing, enabling engineers to convert continuous audio, sensor data, or RF signals into discrete digital form while preserving important frequency information. This bridge between continuous and discrete domains is essential in audio processing, image filtering, and communications.

The DE: Second-order filter: \(a_0\ddot{y} + a_1\dot{y} + a_2 y = b_0\ddot{u} + b_1\dot{u} + b_2 u\), discretized to difference equation \(y[n] = -a_1 y[n-1] - a_2 y[n-2] + b_0 u[n] + b_1 u[n-1] + b_2 u[n-2]\).

What to Investigate
  • Discretization methods: how bilinear transforms and other techniques convert continuous-time transfer functions to discrete-time equivalents
  • Frequency response preservation: why careful discretization preserves magnitude and phase response across the audio band
  • Stability in discretization: how step size (sampling rate) affects whether a stable analog filter remains stable after digitization
  • Practical applications: design of anti-aliasing filters before analog-to-digital conversion, equalizers in audio processing, and sensor conditioning

Presentation tip: Show frequency response plots (magnitude and phase) for an analog filter design, then demonstrate how different discretization methods and sampling rates affect the digital filter response, highlighting when aliasing or instability occurs.

CS-04

Neural ODEs — Deep Learning Meets Differential Equations

Ch 1: 1st-Order Advanced Conceptual
+
Background

Neural ODEs represent a paradigm shift in deep learning: instead of stacking discrete neural network layers, researchers now use the output of an ODE solver as the network. This approach models data transformations as continuous processes, improving memory efficiency, enabling adaptive computation, and creating new connections between neural networks and physics-based modeling. This emerging field is reshaping how AI systems are designed and trained.

The DE: \(\frac{d\mathbf{h}(t)}{dt} = f(\mathbf{h}(t), t; \theta)\), where \(\mathbf{h}(t)\) is the hidden state, \(f\) is a learned function parameterized by \(\theta\), and the output is computed by integrating the ODE from \(t=0\) to \(t=T\).

What to Investigate
  • Continuous vs. discrete representations: why viewing data transformations as continuous ODE solutions can improve model expressivity and efficiency
  • Adjoint sensitivity: how backpropagation works through ODE solvers using the adjoint method, enabling gradient computation without storing hidden states
  • Adaptive computation: how ODEs naturally allow variable-depth processing—difficult problems use more computational steps, easy ones use fewer
  • Physics-informed neural networks: how incorporating differential equations into neural networks enables them to learn from physical laws and generalize better

Presentation tip: Compare standard neural networks (discrete layers) with Neural ODEs on a simple classification or regression problem, highlighting the continuous nature of the ODE solution and how it enables adaptive computation depth.

CS-05

Epidemiological Modeling — SIR Model and Virus Spread

Ch 4: Systems Moderate Conceptual + Simulation
+
Background

The SIR (Susceptible-Infected-Recovered) model is a system of coupled first-order ODEs that predicts how infectious diseases spread through populations. During the COVID-19 pandemic, variants of this model guided public health decisions affecting billions of people, informing when to implement lockdowns, vaccinate populations, and lift restrictions. This powerful yet simple model demonstrates how differential equations directly save lives through evidence-based policy.

The DE: \(\frac{dS}{dt} = -\beta SI\), \(\frac{dI}{dt} = \beta SI - \gamma I\), \(\frac{dR}{dt} = \gamma I\), where \(S\) is susceptible, \(I\) is infected, \(R\) is recovered, \(\beta\) is transmission rate, and \(\gamma\) is recovery rate.

What to Investigate
  • Basic reproduction number \(R_0\): the critical threshold determining whether a disease dies out or causes an epidemic, and how vaccination lowers \(R_0\)
  • Disease dynamics: the interplay between infection waves, herd immunity, and why flattening the curve (reducing peak) is crucial for hospital capacity
  • Parameter estimation: how data from case counts, hospitalizations, and deaths are used to fit the model and make predictions
  • Policy implications: how model predictions guided real-world decisions on lockdowns, quarantine periods, and vaccination strategies during COVID-19

Presentation tip: Build an interactive SIR simulator where users can adjust \(\beta\), \(\gamma\), and initial conditions, watching how the epidemic curve changes. Compare model predictions with real COVID-19 data to show practical accuracy and relevance.

CS-06

Thermal Management in Data Centers — Cooling Computational Infrastructure

Ch 1: 1st-Order Moderate Conceptual + Simulation
+
Background

Heat dissipation in data centers follows Newton's Law of Cooling—a first-order ODE relating temperature rise to power consumption and cooling capacity. As computational demands increase, thermal management becomes the dominant cost in data centers, sometimes exceeding the cost of the actual computing hardware. Engineers use these differential equations to design cooling systems, predict temperature hotspots, and optimize server placement to minimize energy waste.

The DE: \(m c_p \frac{dT}{dt} = P - hA(T - T_{amb})\), where \(m\) is mass of server hardware, \(c_p\) is heat capacity, \(P\) is power dissipation, \(h\) is convection coefficient, \(A\) is surface area, and \(T_{amb}\) is ambient temperature.

What to Investigate
  • Heat balance equation: how power dissipation and cooling capacity must balance to maintain safe operating temperatures
  • Thermal time constant: how quickly a server heats up and cools down, and why this matters for dynamic load management
  • Spatial temperature gradients: how hot spots form in data centers due to non-uniform airflow and unequal power distribution
  • Cooling strategies: liquid cooling, hot-aisle containment, economizers that use outside air, and how these reduce total data center energy cost (PUE)

Presentation tip: Simulate a row of servers with varying power loads and different cooling strategies, showing how temperature evolves and how strategic placement and cooling design can reduce energy waste by 30-50%.

CS-07

Autonomous Vehicle Path Planning — Trajectory Optimization Using DEs

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

Autonomous vehicles must continuously plan smooth, safe trajectories that respect vehicle dynamics, which are inherently described by systems of differential equations. These trajectory planners must generate paths that avoid collisions with obstacles and other vehicles while respecting acceleration and steering limits. The sophistication of DE-based path planning directly determines whether self-driving cars can navigate complex urban environments safely and smoothly.

The DE: System of ODEs: \(\dot{x} = v\cos\theta\), \(\dot{y} = v\sin\theta\), \(\dot{\theta} = (v/L)\tan\phi\), where \((x,y)\) is position, \(\theta\) is heading, \(v\) is speed, \(\phi\) is steering angle, and \(L\) is wheelbase.

What to Investigate
  • Kinematic constraints: why vehicles cannot move laterally or turn on a dime—the differential equations encode fundamental physical limits
  • Trajectory planning algorithms: how planners use these equations as constraints to generate smooth, collision-free paths in real time
  • Control feedback: how the vehicle's control system (steering and throttle) continuously corrects deviations from the planned trajectory
  • Complex scenarios: how DE-based planners handle interactions with other vehicles, pedestrians, and dynamic obstacles in urban environments

Presentation tip: Create a simulator showing a vehicle navigating an urban environment with obstacles, other vehicles, and pedestrians. Demonstrate how the planning algorithm uses the vehicle dynamics equations to generate collision-free trajectories and how feedback control keeps the vehicle on track.

CS-08

Blockchain and Consensus — Differential Equation Models of Network Agreement

Ch 1: 1st-Order Advanced Conceptual
+
Background

Distributed systems must reach consensus (agreement on state) despite faulty nodes, communication delays, and adversarial actors. Differential equations model how opinion evolves through a network as nodes communicate with neighbors, revealing why some consensus algorithms converge quickly while others are vulnerable to manipulation. This emerging field uses dynamical systems theory to prove correctness and security properties of blockchain and distributed ledger systems.

The DE: \(\dot{x}_i = \sum_{j \in \mathcal{N}_i} (x_j - x_i)\), where \(x_i\) is the state (vote/proposal) of node \(i\) and \(\mathcal{N}_i\) is the set of neighbors node \(i\) communicates with.

What to Investigate
  • Convergence conditions: under what network topologies and communication delays does consensus occur, and how fast does it converge
  • Byzantine robustness: how to design algorithms that reach consensus even when some nodes are faulty or controlled by adversaries
  • Proof-of-work vs. proof-of-stake: how differential equation analysis reveals trade-offs in security, energy efficiency, and finality
  • Network effects: how the topology and connectivity of the network affect the speed and stability of consensus algorithms
🏭

Industrial Engineering

— 8 project ideas
IE-01

Inventory Management — The EOQ Model with Continuous Review

Ch 1: 1st-Order Introductory Conceptual
+
Background

The Economic Order Quantity (EOQ) model uses first-order differential equations to balance inventory holding costs against ordering costs. Most business inventory follows a sawtooth pattern: high after an order arrives, then steadily depletes as customers purchase, triggering a reorder when stock falls below a threshold. This model is foundational in supply chain management, enabling companies to minimize waste while avoiding stockouts that lose sales.

The DE: \(\frac{dI}{dt} = -D\), where \(I(t)\) is inventory level and \(D\) is constant demand rate; the cost optimization leads to the classical EOQ formula \(Q^* = \sqrt{2DS/h}\).

What to Investigate
  • Trade-offs between holding and ordering costs: why larger, less frequent orders aren't always optimal
  • Reorder point calculation: why safety stock is needed when demand is uncertain, and how to set reorder points to balance stock-out risk against carrying costs
  • Just-in-time manufacturing: how EOQ principles guide strategy toward minimal inventory and frequent deliveries
  • Multi-item inventory: how practical constraints (storage space, ordering capacity) change the optimal strategy
IE-02

Quality Control and Reliability — The Bathtub Curve

Ch 1: 1st-Order Moderate Conceptual
+
Background

The bathtub curve models how failure rate changes over a product's lifetime: high during early burn-in, low and constant during useful life, then rising again as wear dominates. This curve, derived from differential equations governing failure probability, guides maintenance scheduling, warranty periods, and replacement decisions across industries. Understanding the curve enables intelligent decisions about when to replace items before they fail catastrophically.

The DE: \(\frac{dR(t)}{dt} = -h(t)R(t)\), where \(R(t)\) is reliability (probability of no failure by time \(t\)) and \(h(t)\) is the hazard (failure) rate; the bathtub shape reflects a composite hazard function combining infant mortality, useful-life failures, and wear-out.

What to Investigate
  • Infant mortality: why new components sometimes fail quickly and how burn-in testing reduces early failures
  • Weibull distribution: the family of distributions that capture bathtub behavior and how shape parameter relates to failure mechanisms
  • Maintenance strategies: condition-based maintenance (monitoring for wear) vs. preventive maintenance (fixed schedules based on bathtub curve)
  • Mean time to failure (MTTF): how to calculate expected product life and set warranty periods to balance customer satisfaction against warranty costs

Presentation tip: Show empirical failure data from real products (e.g., electronics, automotive components) and fit them to different sections of the bathtub curve, demonstrating how the model captures real-world reliability behavior and guides maintenance decisions.

IE-03

Supply Chain Dynamics — The Bullwhip Effect

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

The bullwhip effect describes how small fluctuations in consumer demand get amplified into wild swings in orders upstream through the supply chain. A retailer's order variability causes distributors to order even more variably, who cause manufacturers to swing production wildly, leading to excess inventory, stockouts, and waste. This phenomenon, modeled by systems of differential equations, explains why supply chains are inherently unstable and how information sharing can stabilize them.

The DE: System of ODEs for echelon inventory: \(\dot{I}^{(i)} = O^{(i-1)} - D^{(i)}\), where \(I^{(i)}\) is inventory at stage \(i\), \(O^{(i-1)}\) is orders received from upstream, and \(D^{(i)}\) is demand passed downstream; ordering policies introduce feedback and amplification.

What to Investigate
  • Order amplification: why a 10% demand increase can trigger 30-50% swings in manufacturer production, and the mechanisms driving this
  • Information sharing: how sharing actual retail sales data (instead of only orders) dampens oscillations and reduces bullwhip
  • Time delays: how lead times in manufacturing and transportation amplify the effect and destabilize the supply chain
  • Practical mitigation strategies: vendor-managed inventory, continuous replenishment, and collaborative forecasting used by companies like Walmart

Presentation tip: Build a multi-stage supply chain simulator (retail → distributor → manufacturer) with adjustable lead times and ordering policies. Show how small demand fluctuations amplify upstream, then demonstrate how information sharing and better ordering strategies stabilize the system.

IE-04

Chemical Reactor Design — Reaction Kinetics

Ch 1: 1st-Order Moderate Conceptual
+
Background

Chemical reaction rates are governed by differential equations describing how reactant concentrations change over time. Engineers use rate laws (first-order, second-order, zero-order) to predict conversion in batch and continuous reactors, guide temperature and pressure control, and optimize reactor geometry. This foundational knowledge determines everything from pharmaceutical manufacturing to large-scale chemical production.

The DE: First-order: \(\frac{dC_A}{dt} = -k C_A\); Second-order: \(\frac{dC_A}{dt} = -k C_A^2\), where \(C_A\) is reactant concentration, \(k(T)\) is temperature-dependent rate constant following Arrhenius law \(k = Ae^{-E_a/RT}\).

What to Investigate
  • Reaction orders: how to determine if a reaction is first-order, second-order, or more complex from experimental data
  • Temperature dependence: why even small changes in temperature dramatically affect reaction rate through the Arrhenius equation
  • Batch vs. continuous reactors: how to use rate equations to size reactors for desired conversion and throughput
  • Exothermic reactions and safety: how runaway reactions develop (accelerating rates due to heat release) and how to control them

Presentation tip: Show experimental concentration-vs-time data for different reaction orders, fit them to corresponding ODEs, and demonstrate how the rate constant changes with temperature, illustrating the Arrhenius relationship.

IE-05

Queuing Theory and Service Systems — Waiting Line Models

Ch 1: 1st-Order Moderate Conceptual + Simulation
+
Background

Queuing theory models waiting lines in hospitals, factories, and service centers using first-order differential equations (birth-death processes). These models predict waiting times, queue lengths, and server utilization, enabling managers to balance customer satisfaction (short waits) against operational cost (number of servers). Hospitals use queuing models to staff emergency departments, factories use them to schedule maintenance crews, and call centers use them to staff customer service.

The DE: M/M/c queue: \(\frac{dP_n}{dt} = \lambda P_{n-1} - (\lambda + n\mu)P_n + (n+1)\mu P_{n+1}\) for \(n < c\), and modified for \(n \geq c\) where \(c\) is the number of servers.

What to Investigate
  • Server utilization vs. waiting time: why slightly adding capacity dramatically reduces waiting time near saturation
  • Steady-state metrics: average queue length, average waiting time, probability of waiting, derived directly from the ODE steady-state solution
  • Optimal staffing: how to balance the cost of additional servers against the cost of customer dissatisfaction from long waits
  • Applications across industries: hospital emergency departments, factory maintenance, call center staffing, and bank teller allocation

Presentation tip: Simulate a service system (e.g., hospital ER) with adjustable arrival rate and number of servers, showing how queue length and average wait time change. Demonstrate the dramatic nonlinear effect as utilization approaches capacity.

IE-06

Population Dynamics and Resource Management — Logistic Growth

Ch 1: 1st-Order Introductory Conceptual + Simulation
+
Background

The logistic equation models how populations grow under limited resources, reaching a stable carrying capacity. Unlike unbounded exponential growth, the logistic model captures the reality that finite resources (food, space, nutrients) eventually limit population expansion. This differential equation is fundamental to ecology, fisheries management, and conservation biology, informing decisions about sustainable harvesting and species preservation.

The DE: \(\frac{dP}{dt} = rP\left(1 - \frac{P}{K}\right)\), where \(P\) is population, \(r\) is intrinsic growth rate, and \(K\) is carrying capacity.

What to Investigate
  • Growth phases: exponential growth when population is small, then slowing as resources become scarce, finally stabilizing at carrying capacity
  • Carrying capacity: the maximum population sustainable by available resources, and how this reflects environmental limits
  • Sustainable harvesting: how much a population can be harvested annually without decline, and why overfishing leads to collapse
  • Applications to fisheries: why some fish stocks recover while others collapse, and how quotas derived from logistic models maintain productivity

Presentation tip: Show simulation of fish population growth with different harvesting rates, demonstrating sustainable yield at one rate and population collapse at a slightly higher rate, illustrating the critical importance of proper management.

IE-07

Project Scheduling and Learning Curves

Ch 1: 1st-Order Moderate Conceptual
+
Background

Learning curves model how productivity improves with cumulative experience using a power-law ODE that captures the decreasing rate of improvement. Manufacturing productivity increases as workers and processes become familiar with tasks, but the improvement rate slows over time. This differential equation model enables project managers to predict labor costs accurately and schedule manufacturing projects realistically, accounting for the natural learning process.

The DE: Wright's learning curve: \(\frac{dT}{dN} = -a N^{-b}\), where \(T\) is time per unit, \(N\) is cumulative units produced, and \(a, b\) are constants (typically \(b \approx 0.3\)); solution gives \(T(N) = a N^{-b}\).

What to Investigate
  • Improvement rate: why early units take much longer than later units, and how to quantify this through learning curve exponents
  • Learning curve effects in manufacturing: how the rate of cost reduction depends on production volume and is critical for profitability
  • Project scheduling implications: why initial production runs are slow and costly, informing realistic timelines and budget allocations
  • Cumulative vs. individual experience: how learning curves apply differently to individual workers vs. entire production lines and organizations
IE-08

Energy Systems Optimization — Battery Charging and Degradation

Ch 1: 1st-Order Advanced Conceptual + Simulation
+
Background

Battery dynamics during charging and degradation are governed by coupled first-order ODEs modeling state of charge, temperature, and capacity fade. Electric vehicle fleets, power grids with renewable energy storage, and consumer electronics all depend on accurate battery models to maximize efficiency, extend lifespan, and ensure safe operation. The differential equations reveal why charging rates must be limited (to prevent thermal runaway), why cold weather reduces range, and how to extend battery life.

The DE: State of Charge: \(\frac{dSOC}{dt} = \frac{I(t)}{Q_{\max}}\); Temperature: \(\frac{dT}{dt} = \frac{R I^2 - h(T-T_{amb})}{mc_p}\); Capacity fade: \(\frac{dQ}{dt} = -k_{fade}(T) \cdot f(SOC, I)\), where \(I\) is charging current, \(R\) is resistance, \(h\) is heat transfer coefficient.

What to Investigate
  • State of Charge (SOC) dynamics: how charging current affects the rate at which battery fills, and why constant-current/constant-voltage charging is used
  • Thermal effects: why fast charging generates heat, how this accelerates degradation, and the trade-off between speed and longevity
  • Capacity fade mechanisms: why batteries lose capacity over cycles, and how temperature, depth of discharge, and charging rate affect lifespan
  • Fleet management optimization: how to schedule charging in EV fleets to balance range needs against battery longevity and energy cost

Presentation tip: Simulate EV charging scenarios with different charging rates, ambient temperatures, and operating patterns, showing how temperature rises, how fast charging reduces lifespan, and how smart charging strategies extend battery life while meeting transportation needs.

🤖

Mechatronics

— 8 project ideas
MX-01

PID Control of a Robotic Arm

Ch 2: 2nd-Order Moderate Conceptual + Simulation
+
Background

Robotic arms must move smoothly to precise positions while resisting external disturbances. A proportional-integral-derivative (PID) controller uses feedback from the current position to continuously adjust motor commands, modeled as a second-order system with damping and stiffness terms. Without differential equations, engineers cannot predict whether a robot will oscillate, overshoot its target, or respond too slowly to commands.

The DE: \(m\ddot{x} + c\dot{x} + kx = F(t)\) where the control force \(F(t)\) depends on position error and its derivatives, creating stable feedback.

What to Investigate
  • How does each PID term (proportional, integral, derivative) affect response speed and stability?
  • Why do robots require "tuning" — what are the limits of a fixed controller across different payloads?
  • How do industrial robots avoid oscillation when picking up parts of unknown mass?
  • What role do second-order dynamics play in preventing jerky, inefficient motion?

Presentation tip: Create a simulation comparing under-damped, critically-damped, and over-damped responses. Show how PID gains tune the system from oscillatory to smooth motion.

MX-02

Sensor Fusion and Kalman Filtering

Ch 1: 1st-Order Advanced Conceptual + Simulation
+
Background

Self-driving cars combine measurements from GPS, radar, lidar, and cameras, each corrupted by noise and uncertainty. The Kalman filter solves a system of first-order differential equations to estimate the true state (position, velocity) from noisy measurements and a physical model of motion. This technique is essential because no single sensor is perfect, but a differential equation can optimally blend noisy information into an accurate state estimate.

The DE: \(\dot{\mathbf{x}} = \mathbf{A}\mathbf{x} + \mathbf{w}(t)\) where \(\mathbf{x}\) is the state (position, velocity), and \(\mathbf{w}(t)\) represents process noise. The Kalman filter recursively minimizes estimation error using the differential equation as the motion model.

What to Investigate
  • Why do self-driving cars need both a motion model (DE) and measurement sensors — what happens if you rely only on one?
  • How does the Kalman filter balance trust in the model versus trust in measurements?
  • What role does the first-order ODE play in predicting where an object will be in the next time step?
  • How do autonomous vehicles use state estimation for collision avoidance and path planning?

Presentation tip: Simulate a vehicle tracking scenario with GPS and radar noise. Show how the Kalman filter produces a smooth trajectory compared to raw sensor data, then explain how the underlying DE predicts future state.

MX-03

Electric Vehicle Battery Management

Ch 1: 1st-Order Moderate Conceptual + Simulation
+
Background

An electric vehicle's range depends on accurately estimating the state of charge (SOC) of its battery pack, which cannot be measured directly. Engineers model the battery's discharge as a first-order differential equation relating current draw to the rate of charge depletion. Without this differential equation, drivers would not know whether their vehicle can reach the next charging station.

The DE: \(\frac{dQ}{dt} = -I(t)\) where \(Q(t)\) is the available charge (Ah) and \(I(t)\) is the current drawn. The state of charge is \(\text{SOC} = Q(t)/Q_{\max}\), predicting range from the DE solution.

What to Investigate
  • Why is range estimation harder than it sounds — how do temperature, aging, and driving style affect battery dynamics?
  • How do EV battery management systems use this ODE to prevent over-discharge and over-charge?
  • What role does the rate of discharge play in determining remaining range?
  • How can understanding the battery DE improve efficiency and battery longevity?

Presentation tip: Compare predicted SOC (from integrating the ODE) with actual remaining charge under different driving scenarios. Show how real batteries deviate from simple models and why engineers use more complex versions.

MX-04

Inverted Pendulum — The Segway Problem

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

A Segway balances on two wheels while standing upright — a fundamentally unstable configuration. The inverted pendulum system couples the angle of the pole and the position of the cart through a system of nonlinear differential equations. By linearizing around the unstable equilibrium and applying feedback control, engineers make the system stable. This is the core principle behind all balancing robots and self-balancing personal transporters.

The DE: A system of two coupled ODEs: \(\ddot{\theta} = \frac{g}{L}\sin(\theta) + \frac{1}{mL}\cos(\theta)a(t)\) and \(\ddot{x} = a(t) + f(\theta, \dot{\theta})\), where feedback control \(a(t)\) depends on angle and angular velocity to maintain balance.

What to Investigate
  • Why is the inverted pendulum inherently unstable — what happens if the control is turned off?
  • How does linearization around the upright position create a stable control law for small disturbances?
  • What role do sensor delays and actuator time constants play in stability?
  • How do Segways and quadrupedal robots apply these principles to real-world balancing?

Presentation tip: Simulate the nonlinear system and the linearized approximation side-by-side. Show how feedback control (based on the DE) recovers balance from small perturbations, and how open-loop (no control) leads to inevitable failure.

MX-05

Pneumatic and Hydraulic Actuator Dynamics

Ch 2: 2nd-Order Moderate Conceptual
+
Background

Pneumatic and hydraulic actuators power industrial automation, from manufacturing robots to construction equipment. These systems involve fluid dynamics coupled with mechanical motion, creating first- and second-order differential equations. Understanding these ODEs is critical for predicting response time, stability, and efficiency in automated systems.

The DE: \(m\ddot{x} + b\dot{x} = P_1 A_1 - P_2 A_2\) for the mechanical side, coupled with \(\frac{dP}{dt} = \frac{\beta}{V}(Q_{\text{in}} - Q_{\text{out}})\) for fluid compressibility, where pressure \(P\) drives piston acceleration through area \(A\).

What to Investigate
  • How does fluid compressibility introduce damping and oscillation in pneumatic systems?
  • Why are hydraulic actuators faster and more responsive than pneumatic ones?
  • What role do valve dynamics play in controlling actuator response time?
  • How do industrial engineers tune pneumatic systems to avoid instability and vibration?
MX-06

Drone Flight Dynamics — Quadrotor Stability

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

A quadrotor drone must simultaneously control its thrust, roll, pitch, and yaw while remaining stable in the air. The dynamics are described by a system of six coupled nonlinear differential equations governing position and orientation in 3D space. Real-time solving of these equations is essential for onboard flight controllers to stabilize the drone against wind and maintain desired trajectories.

The DE: A system of 12 first-order ODEs (or 6 second-order): \(m\ddot{\mathbf{r}} = -mg\hat{z} + \mathbf{F}_{\text{thrust}}\) for translation, and \(\mathbf{I}\ddot{\boldsymbol{\theta}} = \boldsymbol{\tau}\) for rotation, where thrust and torque depend on rotor speeds controlled by feedback.

What to Investigate
  • How do quadrotors achieve altitude hold using only differential equations and propeller thrust?
  • Why must flight controllers solve these equations in real-time (often 100+ Hz)?
  • What role does linearization play in designing stable attitude and position controllers?
  • How do wind disturbances and measurement delays affect drone stability?

Presentation tip: Use a drone simulator (Gazebo, CopterSim) to show real-time DE solving. Visualize how the flight controller adjusts rotor speeds based on accelerometer and gyroscope feedback to maintain stability and execute commands.

MX-07

Stepper Motor Positioning — Open-Loop vs Closed-Loop

Ch 2: 2nd-Order Moderate Conceptual + Simulation
+
Background

Stepper motors position the print head in 3D printers and CNC machines by stepping through discrete angular positions. However, each step involves complex electromagnetic and mechanical dynamics modeled as a second-order differential equation with natural resonance. Understanding this resonance is crucial to achieving fast, accurate positioning without missed steps or oscillation.

The DE: \(J\ddot{\theta} + b\dot{\theta} + k\theta = \tau_{\text{coil}}\) where the coil torque \(\tau_{\text{coil}}\) is applied in discrete steps. The rotor's moment of inertia \(J\), damping \(b\), and spring constant \(k\) determine whether the motor smoothly reaches target positions or overshoots and oscillates.

What to Investigate
  • Why do stepper motors sometimes miss steps when accelerated too quickly?
  • How does resonance occur between step frequency and the motor's natural frequency?
  • What is the tradeoff between speed and accuracy in open-loop control?
  • How do closed-loop steppers (with feedback) improve print quality and positioning precision?

Presentation tip: Compare open-loop (constant step rate) vs closed-loop (feedback-adjusted) stepping using frequency response plots. Show how matching step frequency to the motor's natural frequency causes resonance and failures.

MX-08

Smart Prosthetics and Biomechanical Modeling

Ch 2: 2nd-Order Advanced Conceptual + Simulation
+
Background

Modern prosthetic limbs use motors and sensors to restore natural motion to people with amputations. A motorized prosthetic leg must model the dynamics of the biological limb it replaces — how forces from muscles and tendons accelerate the limb through space. This is captured by second-order differential equations that describe mass, damping, and stiffness, combined with feedback control to synchronize prosthetic motion with user intent.

The DE: \(m\ddot{x} + c\dot{x} + kx = F_{\text{motor}}(t) + F_{\text{user}}\) where the motor force responds to electromyography (EMG) signals from the residual limb, enabling the prosthesis to move naturally.

What to Investigate
  • How do prosthetics measure user intent from EMG signals and translate it to motor commands?
  • Why is accurate limb dynamics crucial for natural, energy-efficient walking?
  • What role does feedback control play in synchronizing prosthetic motion with the user's movement?
  • How can differential equations improve quality of life for prosthetic users?

Presentation tip: Demonstrate a prosthetic simulator showing how different damping and stiffness parameters affect gait stability. Explain how DEs enable prosthetics to adapt to different walking speeds and terrains.

📐

Mathematics

— 8 project ideas
MA-01

Existence and Uniqueness — When Do Solutions Exist?

Ch 1: 1st-Order Advanced Analytical
+
Background

Not every differential equation has a solution, and some have infinitely many solutions. The Picard-Lindelöf theorem provides precise conditions on the function \(f(t, y)\) in \(\frac{dy}{dt} = f(t, y)\) that guarantee a unique solution exists in a neighborhood of an initial condition. Understanding these conditions is fundamental to applied mathematics, as it tells engineers whether their models are well-posed and predictable.

The DE: Consider \(\frac{dy}{dt} = f(t, y)\). If \(f\) and \(\frac{\partial f}{\partial y}\) are continuous in a rectangle around the initial condition, then a unique solution exists locally (Picard-Lindelöf theorem).

What to Investigate
  • What happens at points where the Picard-Lindelöf conditions fail — can multiple solutions emerge?
  • How does the classical example \(\frac{dy}{dt} = \sqrt{|y|}\) with \(y(0) = 0\) violate uniqueness?
  • Why is uniqueness important for predicting physical phenomena — what goes wrong if solutions are not unique?
  • How do global existence results extend local existence guarantees to longer time intervals?

Presentation tip: Construct explicit examples where existence and uniqueness fail. Visualize non-unique solutions and explain how small perturbations to \(f\) restore uniqueness.

MA-02

Phase Portraits and Qualitative Analysis

Ch 4: Systems Moderate Conceptual + Simulation
+
Background

Many differential equations cannot be solved in closed form, yet their long-term behavior can be understood through phase portraits — graphical depictions of solution trajectories in the state space. Qualitative analysis classifies equilibria (stable nodes, saddles, spirals) without solving the equations, revealing fundamental insights about system behavior. This approach is invaluable for understanding population dynamics, predator-prey systems, and mechanical oscillations.

The DE: For a system \(\frac{d\mathbf{x}}{dt} = \mathbf{f}(\mathbf{x})\), equilibria satisfy \(\mathbf{f}(\mathbf{x}^*) = \mathbf{0}\). Stability is determined by the Jacobian matrix \(\mathbf{J} = \frac{\partial \mathbf{f}}{\partial \mathbf{x}}\) evaluated at equilibria, without requiring explicit solutions.

What to Investigate
  • How do eigenvalues of the Jacobian determine whether trajectories approach, spiral around, or leave equilibria?
  • What is a separatrix, and why are these boundaries critical for understanding system behavior?
  • How do limit cycles (closed orbits that attract nearby trajectories) emerge in nonlinear systems?
  • How can phase portraits predict long-term population sizes without solving predator-prey equations?

Presentation tip: Create interactive phase portrait visualizations for classic systems (predator-prey, van der Pol oscillator). Show how eigenvalues and eigenvectors of the Jacobian determine the shape of trajectories near equilibria.

MA-03

Fourier Series and the Heat Equation

Ch 3: Laplace Transform Advanced Analytical
+
Background

The heat equation is a fundamental partial differential equation describing temperature diffusion through materials. Separation of variables reduces it to infinitely many ODEs, each solved by exponential decay. Fourier series decompose the initial temperature distribution into periodic harmonics, each of which decays at a different rate. This elegant method connects classical analysis, harmonic analysis, and applied physics.

The DE: \(\frac{\partial u}{\partial t} = k\frac{\partial^2 u}{\partial x^2}\) (the heat equation). Using separation of variables \(u(x,t) = X(x)T(t)\), each mode \(T_n(t) = e^{-\lambda_n^2 kt}\) decays exponentially, with decay rate depending on the frequency of the spatial oscillation.

What to Investigate
  • Why do high-frequency temperature oscillations decay faster than low-frequency ones?
  • How does the Fourier series expansion connect the initial temperature profile to the solution at all later times?
  • What is the physical interpretation of the infinite sum of decaying exponentials?
  • How do engineers use this principle to model cooling of electronic devices and heat transfer in buildings?

Presentation tip: Animate the heat equation solution, showing how Fourier modes decay at different rates. Start with a non-smooth initial condition (square wave) and illustrate how it smooths over time as high frequencies vanish first.

MA-04

Chaos Theory — The Lorenz System

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

The Lorenz system is a simplified atmospheric convection model consisting of three coupled ODEs. Despite being deterministic, tiny differences in initial conditions lead to vastly different trajectories — the "butterfly effect." The Lorenz attractor exhibits chaotic behavior while remaining confined to a bounded region of phase space. This discovery revolutionized our understanding of long-term weather prediction and dynamical systems.

The DE: \(\frac{dx}{dt} = \sigma(y - x)\), \(\frac{dy}{dt} = x(\rho - z) - y\), \(\frac{dz}{dt} = xy - \beta z\), where \(\sigma\), \(\rho\), and \(\beta\) are parameters. For certain values (e.g., \(\rho = 28\)), the system exhibits chaos.

What to Investigate
  • What makes the Lorenz system chaotic — why do small differences grow exponentially?
  • How does the strange attractor constrain trajectories while remaining sensitive to initial conditions?
  • What is the butterfly effect, and why does it limit weather prediction to roughly two weeks?
  • How do Lyapunov exponents quantify the rate of divergence between nearby trajectories?

Presentation tip: Simulate the Lorenz system for two very close initial conditions and display the trajectories side-by-side. Show how they start nearly identical but diverge dramatically. Visualize the strange attractor in 3D.

MA-05

Numerical Methods — When Exact Solutions Don't Exist

Ch 1: 1st-Order Moderate Analytical + Simulation
+
Background

Most real-world differential equations lack closed-form solutions and must be solved numerically. Methods like Euler's method and Runge-Kutta advance the solution one step at a time by approximating the slope. Understanding how numerical errors accumulate is critical for engineering applications: satellite trajectories, climate models, and drug delivery simulations depend on numerical DE solvers with controlled accuracy.

The DE: \(\frac{dy}{dt} = f(t, y)\) with initial condition \(y(t_0) = y_0\). Numerical methods approximate \(y(t_{n+1}) \approx y(t_n) + h \cdot \Phi(t_n, y_n, h)\), where \(\Phi\) is a slope estimator and \(h\) is the step size.

What to Investigate
  • How does step size affect accuracy and computational cost — what is the optimal tradeoff?
  • Why do higher-order methods (Runge-Kutta) produce better accuracy than Euler's method?
  • What is local truncation error versus global error, and how do they relate?
  • How do adaptive step-size methods ensure accuracy while minimizing computation?

Presentation tip: Implement Euler and Runge-Kutta methods on a simple ODE with a known analytical solution. Plot both the numerical and exact solutions, showing how error decreases with smaller step sizes and increases with method order.

MA-06

Differential Equations in Cryptography and Number Theory

Ch 1: 1st-Order Advanced Conceptual
+
Background

Continuous differential equations can approximate discrete number-theoretic phenomena, bridging the gap between calculus and pure mathematics. The prime number theorem, for instance, is proven using complex analysis and differential equations applied to the Riemann zeta function. Elliptic curve cryptography relies on understanding differential equations on algebraic curves, which form the backbone of modern secure communications.

The DE: The logarithmic integral \(\text{li}(x) = \int_0^x \frac{dt}{\ln t}\) approximates the prime counting function \(\pi(x)\). The prime number theorem states \(\lim_{x \to \infty} \frac{\pi(x)}{\text{li}(x)} = 1\), connecting discrete prime distribution to a continuous integral.

What to Investigate
  • How do continuous approximations to discrete functions provide insight into number theory?
  • What role does the Riemann zeta function and its derivatives play in understanding primes?
  • How do elliptic curves (defined by nonlinear equations) enable modern cryptographic systems?
  • Why is differential geometry essential for understanding cryptographic protocols?
MA-07

Dynamical Systems and Bifurcation Theory

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

Bifurcation theory studies how qualitative behavior of dynamical systems changes as parameters vary. A small change in a parameter can trigger a sudden transition: equilibria may appear or disappear, stable limit cycles may emerge, or chaotic behavior may ignite. Understanding bifurcations is essential for predicting critical transitions in ecosystems, mechanical systems, and engineering designs.

The DE: Consider \(\frac{dx}{dt} = f(x; \mu)\) where \(\mu\) is a parameter. A bifurcation occurs at a critical value \(\mu_c\) where the stability or number of equilibria changes qualitatively. For example, a saddle-node bifurcation occurs when two fixed points collide and annihilate.

What to Investigate
  • What types of bifurcations exist (saddle-node, pitchfork, Hopf), and how do they manifest geometrically?
  • How do bifurcations explain sudden transitions in ecological populations or engineering systems?
  • What role does the Jacobian matrix play in predicting bifurcations?
  • How can understanding bifurcations help design systems that avoid unwanted transitions?

Presentation tip: Create bifurcation diagrams showing how equilibria and periodic orbits change with parameter variation. Animate the phase portrait as the parameter sweeps through a bifurcation point, showing qualitative changes in behavior.

MA-08

Stochastic Differential Equations — When Randomness Meets Calculus

Ch 1: 1st-Order Advanced Conceptual + Simulation
+
Background

Real systems are never perfectly deterministic — they are buffeted by noise from thermal fluctuations, measurement uncertainty, and unmodeled disturbances. Stochastic differential equations incorporate randomness explicitly through Brownian motion terms, extending classical ODEs to capture realistic uncertainty. These equations are essential for modeling financial markets, molecular dynamics, and noise-driven phenomena in biology and physics.

The DE: \(dX_t = f(X_t) \, dt + g(X_t) \, dW_t\) where \(W_t\) is a Wiener process (Brownian motion). The term \(g(X_t) \, dW_t\) represents random fluctuations, while \(f(X_t) \, dt\) is the deterministic drift. The solution is a stochastic process, not a deterministic curve.

What to Investigate
  • How does Brownian motion differ from deterministic trajectories — what is the limiting behavior?
  • Why is standard calculus inapplicable to SDEs — what is Ito calculus and why is it necessary?
  • How do SDEs model stock prices in finance (geometric Brownian motion) and molecular motion in physics?
  • What is the relationship between SDEs and their corresponding Fokker-Planck partial differential equations?

Presentation tip: Simulate sample paths of classic SDEs (geometric Brownian motion, Ornstein-Uhlenbeck process). Show an ensemble of trajectories and compute statistics like mean and variance at each time, illustrating how randomness creates a distribution of outcomes.

🧪

Chemical Engineering

— 10 project ideas
CH-01

Chemical Reactor Design — The CSTR and Plug Flow Models

Ch 1: 1st-Order Moderate Conceptual + Simulation
+
Background

Chemical reactors are the heart of every chemical plant. The continuous stirred-tank reactor (CSTR) and plug flow reactor (PFR) are modeled by first-order ODEs derived from mass and energy balances. These differential equations predict how reactant concentrations change with time or position, enabling engineers to size reactors, optimize yield, and ensure safe operation. Understanding these models is fundamental to process design in petroleum refining, pharmaceutical manufacturing, and polymer production.

The DE: \(V\frac{dC_A}{dt} = F(C_{A0} - C_A) - Vk C_A\) for a CSTR, and \(\frac{dC_A}{dV} = \frac{r_A}{F}\) for a PFR, where \(C_A\) is concentration, \(F\) is flow rate, \(k\) is the rate constant, and \(r_A\) is the reaction rate.

What to Investigate
  • How do CSTR and PFR models differ in their mathematical formulation, and what does this mean for reactor performance?
  • What role does residence time play in determining conversion and selectivity?
  • How do engineers use these ODE models to size reactors for desired production rates?
  • What happens when multiple reactions occur simultaneously — how do coupled ODEs describe selectivity?

Presentation tip: Compare CSTR and PFR performance for the same reaction by plotting conversion vs. reactor volume. Show how the ODE solutions reveal which reactor type is more efficient for different reaction orders.

CH-02

Heat Exchangers — Temperature Profiles and Effectiveness

Ch 1: 1st-Order Moderate Conceptual
+
Background

Heat exchangers transfer thermal energy between fluid streams and appear in every chemical plant, power station, and HVAC system. The temperature profiles along a heat exchanger are governed by coupled first-order ODEs derived from energy balances on each fluid stream. These equations predict outlet temperatures, required surface area, and thermal effectiveness — all critical for energy-efficient process design.

The DE: \(\frac{dT_h}{dx} = -\frac{UA}{m_h c_{p,h}}(T_h - T_c)\) and \(\frac{dT_c}{dx} = \frac{UA}{m_c c_{p,c}}(T_h - T_c)\), where \(T_h, T_c\) are hot and cold stream temperatures, \(U\) is the overall heat transfer coefficient, and \(A\) is the exchange area.

What to Investigate
  • How do co-current and counter-current flow arrangements produce different temperature profiles, and why is counter-current more efficient?
  • What does the concept of effectiveness-NTU tell us about heat exchanger performance?
  • How do engineers use these ODE models to determine required exchanger size and material selection?
  • What are real-world challenges like fouling, and how do they alter the DE parameters over time?

Presentation tip: Plot temperature profiles for both co-current and counter-current configurations side by side. Show how the ODE solutions demonstrate the thermodynamic advantage of counter-current flow.

CH-03

Distillation Column Dynamics — Separation of Mixtures

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

Distillation is the most widely used separation process in chemical engineering, from oil refining to alcohol production. Each tray in a distillation column is described by coupled ODEs for mass and energy balances, forming a system of differential equations whose solutions predict composition profiles, required reflux ratios, and column stability. Understanding these dynamics is essential for designing and controlling separation processes.

The DE: \(M_n \frac{dx_n}{dt} = L_{n+1}x_{n+1} + V_{n-1}y_{n-1} - L_n x_n - V_n y_n + Fz_F\delta_n\), a system of ODEs for each tray \(n\), where \(x, y\) are liquid and vapor compositions, \(L, V\) are liquid and vapor flow rates.

What to Investigate
  • How does the system of tray-by-tray ODEs describe the separation of a binary mixture?
  • What is the role of reflux ratio in determining product purity, and how does the DE model predict optimal operation?
  • How do transient disturbances (feed composition changes) propagate through the column according to the coupled ODEs?
  • Why is feedback control essential for distillation, and how do the system dynamics inform controller design?

Presentation tip: Animate the composition profile on each tray over time as the column reaches steady state. Show how changing reflux ratio shifts the solution of the ODE system.

CH-04

Polymer Reaction Kinetics — Chain Growth and Molecular Weight

Ch 1: 1st-Order Advanced Conceptual
+
Background

Polymers — from plastics to synthetic fibers — are created through polymerization reactions governed by systems of ODEs. The evolution of monomer concentration, growing chain populations, and molecular weight distribution over time follows from kinetic rate equations. These differential equation models allow engineers to predict and control the properties of the final polymer product, including average molecular weight, polydispersity, and mechanical strength.

The DE: \(\frac{d[M]}{dt} = -k_p [M][P^*]\) for monomer consumption and \(\frac{d\mu_k}{dt}\) for the \(k\)-th moment of the molecular weight distribution, where \([M]\) is monomer concentration, \([P^*]\) is active polymer concentration, and \(k_p\) is the propagation rate constant.

What to Investigate
  • How do the ODEs for initiation, propagation, and termination determine the final polymer properties?
  • What is molecular weight distribution and why do engineers use moment equations (ODEs) to track it?
  • How does reactor type (batch vs. continuous) affect the polymerization kinetics and product quality?
  • What real-world products depend on precise control of these reaction kinetics?

Presentation tip: Show how monomer conversion and average molecular weight evolve over time by solving the kinetic ODEs. Compare batch and continuous reactor predictions.

CH-05

Diffusion and Mass Transfer — Fick's Law in Practice

Ch 2: 2nd-Order Moderate Conceptual + Simulation
+
Background

Mass transfer by diffusion is central to chemical engineering — from drug delivery through membranes to gas absorption in scrubbers. Fick's second law is a second-order differential equation that describes how concentration changes with position and time. In steady state, it reduces to a second-order ODE whose solutions predict concentration profiles across membranes, catalyst pellets, and biological tissues.

The DE: \(D\frac{d^2C}{dx^2} = 0\) for steady-state diffusion (yielding linear profiles), and \(D\frac{d^2C}{dx^2} - k C = 0\) when a first-order reaction consumes the diffusing species inside a catalyst pellet.

What to Investigate
  • How does the ODE solution predict concentration profiles across membranes and porous catalysts?
  • What is the Thiele modulus and how does it characterize the competition between diffusion and reaction?
  • How do engineers design drug delivery patches, gas separation membranes, and catalytic converters using these models?
  • What happens when diffusion limitations control the overall reaction rate in industrial processes?

Presentation tip: Plot concentration profiles inside a catalyst pellet for different Thiele modulus values. Show how the ODE solution transitions from reaction-limited to diffusion-limited regimes.

CH-06

Process Control — PID Controllers in Chemical Plants

Ch 2: 2nd-Order Advanced Conceptual + Simulation
+
Background

Chemical plants operate safely and efficiently because of automatic control systems. PID (Proportional-Integral-Derivative) controllers are the workhorses of process control, and their behavior is described by second-order ODEs when coupled with process dynamics. Understanding these differential equations reveals why some processes oscillate, how to tune controllers for stability, and what happens during disturbances — knowledge essential for every chemical engineer.

The DE: \(\tau^2\frac{d^2y}{dt^2} + 2\zeta\tau\frac{dy}{dt} + y = K_c\left(e + \frac{1}{\tau_I}\int e\,dt + \tau_D\frac{de}{dt}\right)\), where \(y\) is the process variable, \(e\) is the error signal, and \(K_c, \tau_I, \tau_D\) are controller parameters.

What to Investigate
  • How does the second-order ODE describe the closed-loop response of a controlled chemical process?
  • What do the damping ratio and natural frequency tell us about whether the process will oscillate or settle smoothly?
  • How do engineers tune PID parameters to achieve fast, stable responses without overshooting?
  • What are real examples of process control failures and how do the DEs explain what went wrong?

Presentation tip: Simulate a temperature control loop for a CSTR. Show step responses for underdamped, critically damped, and overdamped cases, and demonstrate the effect of PID tuning.

CH-07

Catalysis and Surface Reactions — The Langmuir-Hinshelwood Model

Ch 1: 1st-Order Advanced Conceptual
+
Background

Catalytic reactions drive much of the chemical industry — from ammonia synthesis to catalytic converters in cars. The Langmuir-Hinshelwood kinetic model describes how molecules adsorb onto catalyst surfaces and react, leading to nonlinear ODEs for surface coverage and gas-phase concentrations. These models are essential for designing catalysts, optimizing reaction conditions, and understanding why some catalysts deactivate over time.

The DE: \(\frac{d\theta}{dt} = k_a C(1-\theta) - k_d \theta - k_r \theta^2\), where \(\theta\) is fractional surface coverage, \(k_a, k_d, k_r\) are adsorption, desorption, and surface reaction rate constants.

What to Investigate
  • How does the nonlinear ODE model capture the competition between adsorption, desorption, and surface reaction?
  • What is catalyst saturation and how does the DE predict when adding more reactant no longer increases the rate?
  • How do automotive catalytic converters, ammonia synthesis (Haber process), and petroleum cracking rely on these models?
  • What causes catalyst deactivation and how can time-dependent DE models describe it?

Presentation tip: Plot reaction rate versus reactant concentration showing the saturation curve predicted by the ODE model. Explain the physical meaning of each regime and connect it to industrial catalyst design.

CH-08

Crystallization Kinetics — Population Balance Equations

Ch 4: Systems Advanced Conceptual + Simulation
+
Background

Crystallization is used in pharmaceutical, food, and chemical industries to produce solid products with specific particle sizes. The crystal size distribution evolves over time according to population balance equations — a system of ODEs that track nucleation (birth of new crystals) and growth rates. These models are critical for producing uniform drug tablets, sugar crystals, and semiconductor materials.

The DE: \(\frac{dn(L,t)}{dt} + G\frac{\partial n}{\partial L} = B\delta(L - L_0)\), simplified to ODE moments: \(\frac{d\mu_j}{dt} = jG\mu_{j-1} + BL_0^j\), where \(n(L,t)\) is the number density of crystals of size \(L\), \(G\) is growth rate, and \(B\) is nucleation rate.

What to Investigate
  • How do the moment ODEs capture the evolution of average crystal size and size distribution over time?
  • What determines whether a crystallizer produces uniform or widely distributed particle sizes?
  • How do pharmaceutical companies use these models to ensure consistent drug tablet properties?
  • What role does supersaturation play in controlling nucleation and growth rates?

Presentation tip: Simulate crystal size distribution evolution in a batch crystallizer. Show how changing cooling rate (which affects supersaturation) shifts the distribution from fine to coarse crystals.

CH-09

Biochemical Engineering — Enzyme Kinetics and Bioreactors

Ch 1: 1st-Order Moderate Conceptual + Simulation
+
Background

Enzyme-catalyzed reactions are the basis of biotechnology — from brewing and baking to pharmaceutical production and biofuels. The Michaelis-Menten model describes enzyme kinetics through a nonlinear ODE that predicts how substrate is consumed over time. Bioreactor design relies on these models to optimize cell growth, product formation, and nutrient feeding strategies in industries ranging from insulin production to wastewater treatment.

The DE: \(\frac{d[S]}{dt} = -\frac{V_{max}[S]}{K_m + [S]}\), where \([S]\) is substrate concentration, \(V_{max}\) is the maximum reaction rate, and \(K_m\) is the Michaelis constant representing enzyme-substrate affinity.

What to Investigate
  • How does the Michaelis-Menten ODE transition between zero-order and first-order kinetics, and what does this mean physically?
  • How do engineers use Lineweaver-Burk plots to extract kinetic parameters from experimental data?
  • What role do bioreactor models (batch, fed-batch, continuous) play in scaling up bioprocesses?
  • How are these models applied in real industries — insulin production, biofuel fermentation, wastewater treatment?

Presentation tip: Solve the Michaelis-Menten ODE numerically and plot substrate depletion curves for different enzyme concentrations. Show how the kinetic parameters determine production efficiency in a bioreactor.

CH-10

Thermal Runaway in Exothermic Reactors — Safety and Stability

Ch 4: Systems Advanced Conceptual
+
Background

Thermal runaway is one of the most dangerous failure modes in chemical plants — when an exothermic reaction generates heat faster than the cooling system can remove it, temperatures escalate exponentially, potentially causing explosions. The coupled mass and energy balance ODEs for a CSTR exhibit multiple steady states and bifurcation behavior. Understanding these dynamics through differential equations is critical for designing safe reactors and preventing industrial disasters.

The DE: \(\frac{dT}{dt} = \frac{(-\Delta H_r)(-r_A)V - UA(T - T_c)}{\rho V c_p}\) coupled with \(\frac{dC_A}{dt} = \frac{F}{V}(C_{A0} - C_A) - k_0 e^{-E_a/RT}C_A\), where the Arrhenius temperature dependence creates nonlinear feedback between temperature and reaction rate.

What to Investigate
  • How do the coupled mass-energy ODEs produce multiple steady states, and what determines which one the reactor operates at?
  • What is the Semenov diagram and how does it visually predict runaway conditions from the heat generation and removal curves?
  • What historical accidents (e.g., Bhopal, T2 Laboratories) were caused by thermal runaway, and how do the DEs explain what happened?
  • How do modern safety systems use these mathematical models to detect and prevent runaway conditions?

Presentation tip: Plot the heat generation and heat removal curves for a CSTR, showing how they intersect to create stable and unstable steady states. Demonstrate how reducing cooling shifts the system into the runaway regime.

📚

Education

— 8 project ideas
ED-01

Teaching Differential Equations Through Real-World Modeling

Ch 1: 1st-Order Moderate Conceptual
+
Background

Traditional DE courses emphasize solving techniques and manipulations, often disconnected from application. Research in mathematics education shows that students develop deeper understanding when they engage with DEs through authentic modeling problems — where they formulate equations from physical principles, not just solve provided ones. This pedagogical approach improves retention and motivates students to persist through technical difficulties.

The DE: Rather than presenting \(\frac{dy}{dt} = -ky\), guided discovery asks: "Why does a cooling coffee cup follow exponential decay?" Students derive \(\frac{dT}{dt} = -k(T - T_{\text{room}})\) from Newton's law of cooling, connecting physical principle to mathematical form.

What to Investigate
  • What role does problem-based learning play in improving conceptual understanding of DEs?
  • How do students progress from formulating differential equations to interpreting solutions?
  • What are the barriers to students transferring DE knowledge from one context to another?
  • How can educators design scaffolded modeling projects that build mathematical sophistication?
ED-02

Visualization Tools for DEs — Slope Fields, Phase Portraits, and Interactive Apps

Ch 4: Systems Moderate Conceptual + Simulation
+
Background

Slope fields and phase portraits transform abstract differential equations into visual landscapes where solutions are trajectories. Interactive tools allow students to explore how initial conditions, parameters, and equation structure shape solution behavior. Visual understanding often precedes symbolic manipulation, making these tools pedagogically powerful for building intuition before rigorous analysis.

The DE: A slope field for \(\frac{dy}{dt} = f(t, y)\) displays short line segments at points \((t, y)\) with slopes equal to \(f(t, y)\). Solutions appear as curves tangent to these segments, visualizing the geometry of the differential equation.

What to Investigate
  • How do interactive visualizations improve student understanding compared to static textbook diagrams?
  • What features make visualization tools effective for teaching: real-time updates, parameter sliders, color coding?
  • How do students use phase portraits to understand stability and long-term behavior without solving equations?
  • What open-source tools (Desmos, GeoGebra, PhET) are best suited for different DE concepts?

Presentation tip: Build interactive Desmos or GeoGebra applets allowing students to adjust parameters and immediately see how slope fields and phase portraits change. Include mini-tutorials showing how to interpret visual features.

ED-03

History of Differential Equations — From Newton to Modern Computing

Ch 1: 1st-Order Introductory Conceptual
+
Background

Differential equations emerged from Newton's attempt to understand motion through calculus. The Bernoulli family, Euler, Lagrange, and others developed solution techniques for increasingly complex equations. Understanding this intellectual history illuminates why certain techniques exist and connects abstract mathematics to the real problems that motivated them. It also humanizes mathematics and reveals how engineering needs drive mathematical innovation.

The DE: Newton's second law, \(m\frac{d^2x}{dt^2} = F(x, \dot{x}, t)\), was the first major differential equation. Its formulation required both the concept of instantaneous rate of change and an understanding of forces — pioneering work in mathematical modeling.

What to Investigate
  • How did Newton's laws of motion lead to the formulation of differential equations?
  • What major breakthroughs (Fourier series, Laplace transforms, numerical methods) changed DE pedagogy and practice?
  • How has computing power transformed which equations we can solve and how we teach DE solutions?
  • How did the practical demands of engineers and physicists drive abstract mathematical developments?

Presentation tip: Create a timeline connecting historical developments in DEs to corresponding applications and discoveries. Show how problems (planetary motion, heat flow, electrical circuits) motivated mathematical innovations.

ED-04

Common Misconceptions in Learning DEs

Ch 1: 1st-Order Moderate Conceptual
+
Background

Research in mathematics education has documented persistent misconceptions that students hold about differential equations: confusing a solution with an equation, treating derivatives as ratios without limits, misinterpreting what "solving" means. Understanding these errors is crucial for educators designing effective instruction. Targeted pedagogical interventions that directly address misconceptions prove more effective than general instruction.

The DE: A common error: if \(\frac{dy}{dt} = 2t\), some students write \(y = t\) (confusing derivative with integration). Proper understanding requires recognizing that \(\frac{dy}{dt}\) is the instantaneous rate of change, not the relationship between \(y\) and \(t\).

What to Investigate
  • What are the most prevalent misconceptions students hold when learning ODEs?
  • Why do students confuse equations with solutions, and how can instruction address this?
  • How do physical intuitions sometimes lead to incorrect mathematical reasoning?
  • What evidence-based instructional strategies successfully remediate misconceptions?

Presentation tip: Develop diagnostic assessments identifying misconceptions. Create targeted interventions (analogies, worked examples, explicit correction) and measure their effectiveness on student learning.

ED-05

Differential Equations in Islamic Golden Age Mathematics

Ch 1: 1st-Order Moderate Conceptual
+
Background

The Islamic Golden Age (8th-15th centuries) produced fundamental mathematical advances that paved the way for differential equations. Al-Khwarizmi's algebra (from which the term "algorithm" derives), Ibn al-Haytham's work on optics and infinitesimals, and developments in trigonometry laid essential groundwork. Recognizing these contributions honors diverse intellectual traditions and shows students that modern mathematics synthesizes knowledge from many cultures.

The DE: Ibn al-Haytham studied how light refracts and reflects, leading to equations describing ray paths. His work on infinitesimals and geometric transformations anticipated calculus concepts essential to differential equations by centuries.

What to Investigate
  • How did algebraic innovations in the Islamic world enable the symbolic manipulations needed for DE solutions?
  • What optical problems did Ibn al-Haytham study, and how do they relate to differential equations?
  • How did mathematical knowledge flow between Islamic scholars, European universities, and other regions?
  • How can curricula better acknowledge diverse contributions to mathematical development?
ED-06

Project-Based Learning in Mathematics — Does It Work?

Ch 4: Systems Moderate Conceptual
+
Background

Project-based learning (PBL) in mathematics asks students to engage with authentic, complex problems over extended periods. Research shows mixed results: well-designed PBL improves conceptual understanding and motivation, but poorly designed projects can overwhelm students and detract from skill development. Understanding when and how PBL is effective for teaching differential equations requires careful investigation of evidence and careful course design.

The DE: Rather than solving textbook problems, a PBL project might ask: "Design a water filtration system — model the pollutant concentration over time using ODEs." Students must formulate, solve, and interpret their own differential equations.

What to Investigate
  • What does research say about PBL effectiveness in mathematics compared to traditional instruction?
  • What characteristics distinguish successful PBL projects from those that overwhelm students?
  • How does PBL affect student motivation, conceptual understanding, and technical skill development?
  • What support structures (scaffolding, formative feedback, checkpoints) make PBL more effective?

Presentation tip: Design and pilot a multi-week PBL module on differential equations. Collect pre/post assessments, student reflections, and engagement data. Compare outcomes to traditional instruction and communicate findings clearly.

ED-07

The Role of Technology in Teaching DEs — From Calculators to AI

Ch 1: 1st-Order Moderate Conceptual
+
Background

Technology has transformed what we teach in differential equations courses. Fifty years ago, solving DEs by hand dominated the curriculum. Computer algebra systems (CAS), numerical solvers, and now AI tools allow students to offload computation and focus on conceptualization and modeling. This shift raises profound questions: what do we teach when computation is automated? How do we maintain mathematical rigor while leveraging technology? Understanding these tensions is essential for modern DE pedagogy.

The DE: A nonlinear ODE like \(\frac{dy}{dt} = y - y^2\) once required sophisticated analytical or numerical techniques. Modern tools solve such equations instantly, allowing students to focus on interpretation: What does the solution curve mean physically? How do equilibria relate to the system's behavior?

What to Investigate
  • How has technology (graphing calculators, MATLAB, Wolfram Alpha, ChatGPT) changed what we teach in DE courses?
  • What mathematical skills remain essential when powerful computational tools are readily available?
  • How can educators design assessments that evaluate understanding rather than computational ability?
  • What are the risks and benefits of AI tools like ChatGPT in mathematics education?

Presentation tip: Compare DE instruction across eras (pre-calculator, calculator era, computer algebra systems, AI). Analyze sample exams and projects to show how curricula have evolved. Discuss what remains constant (conceptual understanding) and what has changed (emphasis on computation).

ED-08

Cross-Disciplinary Teaching — DEs as a Bridge Between Math and Engineering

Ch 4: Systems Moderate Conceptual
+
Background

Mathematics and engineering departments often teach differential equations in isolation, with little communication about how applications motivate theory. Cross-disciplinary collaboration can strengthen both: mathematicians learn what practitioners truly need, engineers gain deeper mathematical insight. Integrated curricula where math and engineering faculty co-teach real problems produce students with both conceptual depth and practical competence.

The DE: An engineer asking "How do I stabilize a control system?" and a mathematician deriving "Lyapunov stability from DE theory" address the same fundamental question. Integrated teaching shows how mathematical and engineering perspectives enrich each other.

What to Investigate
  • What barriers prevent collaboration between mathematics and engineering departments in DE instruction?
  • How can integrated curricula better prepare students for real-world problem-solving?
  • What are examples of successful cross-disciplinary DE courses, and what made them work?
  • How can mathematics gain relevance to engineers while maintaining conceptual rigor?

Presentation tip: Develop a prototype integrated module on a specific topic (e.g., control theory) taught from both mathematical and engineering perspectives. Document student learning outcomes, interdisciplinary understanding, and practical application skills.

Want to Propose Your Own Topic?

The ideas above are suggestions — you are welcome to propose a different topic that connects DEs to your specialty. Just make sure it includes a clear DE model, a solution or simulation, and a connection to real engineering practice. Email your proposal to Dr. Mabrok for approval before starting.