JMU Shell Eco-marathon mathematical model
The JMU Supermileage team has been participating in the Shell Eco-marathon competition since 2018. Shell Eco-marathon is one of the world’s leading student engineering competitions, where STEM students from across the globe design, build and operate some of the world’s most energy-efficient vehicles.
We identified that the enterprise had no strategy to drive around the track nor the platform to analyze the performance of the fuel-efficient vehicle analytically. Our project entails delivering a mathematical model that characterizes the vehicle and helps us make data-driven design decisions by simulating the vehicle around a given track to analyze its performance resourcefully through Simulink.
This project ultimately has helped all of us get an understanding of model-based engineering and the value it brings to us and future JMU supermileage teams! I have been working extensively as a project manager, various sub-team tasks, and also programming our electronic fuel injected system for optimal performance in different operating conditions.
Overall.
-
What
Building a mathematical model of our fuel-efficient vehicle to make data-driven design decisions based on simulations outputted by the model.
-
How
Creating a base concept of a fuel-efficient vehicle and determining design decisions by customizing components code and features as to determine the best output in terms of fuel-efficiency around a raceway.
-
Results
Modeled the drivetrain, engine, and chassis of the fuel-efficient vehicle to improve fuel-efficiency of the vehicle by 26% of the previous year’s competition car e.g. totaling 216 MPGs.
design decisions.
Overall the project, gave me the opportunity to get a taste of early machine learning as the application was to have the vehicle’s subsystems modeled and simulated according to a raceway. The Simulink code would output datapoints of velocity, gas mileage, and data plots for throttle positioning at various points of the track as to determine an optimal strategy around a track.