Design Engineering
Showcase 2020

Eatron Technologies

Student
Mohammed Elfadil
Team
Battery Management System
Supervisor
Dr Michel-Alexandre Cardin
Role
Software and Controls Engineering
Sector
Automotive and Aerospace

The industrial placement was a great opportunity to utilise the various skills we have learnt during our studies and apply them in a commercial context. I spent the last three months working at a software-based automotive start-up called Eatron Technologies. Here I learnt a great deal about machine learning and how it can be utilised most effectively. This internship affirmed my interest in working in the automotive industry. I also believe my deepened understanding of machine learning will be very transferrable in my future projects and professional aspirations, particularly as AI becomes ubiquitous in the future.

— Eatron Technologies

Demonstration of Design Engineering Thinking and Skills

Throughout my placement, I both built on my previous knowledge of machine learning, as well as learn and apply new types of machine learning algorithms. Over the three month timespan, I used three different algorithms to model a batteries Remaining Useful Life (RUL), each with varying accuracies.

— Eatron Technologies
Timeline of entire placement.

The preliminary process for any data-driven project is the collection and preparation of the data. Data was processed using MATLAB and then sent to python before finally being fed into a model. When working with large datasets, in some cases in excess of 50 million samples, it is critical to minimise the time complexity of my code.

— Eatron Technologies

Role and Contributions

Eatron technologies provide intelligent software for both BMS and ADAS. I worked alongside the BMS team based in Warwick. Eatron Technologies’ BMS system already comes equipped with many different capabilities which already make their BMS system competitive in the automotive market. Eatron provides these solutions to some existing low volume premium car manufacturers. My role as a software engineering intern was to use machine learning to develop a model which can accurately predict a battery remaining useful life (RUL). This would be employed directly with some of Eatron’s other BMS capabilities, creating a more competitive product.

— Eatron Technologies

I initially used an elastic net which is a regularization regression algorithm and had high accuracies.

— Eatron Technologies

I then experimented with neural networks of various architectures in an attempt to improve on the accuracies of the elastic net.

— Eatron Technologies

The final model I tried used an RNN which works by finding time dependencies between the data, and ultimately provided the best accuracy of 93%.

— Eatron Technologies

Summary

  • The placement has strongly developed my skills in a wide range of areas, and I felt I have made a strong impact on my place of work.
  • With further development of my technical coding skills, machine learning and time management, I should be in a strong position for future employment opportunities as I continue to grow as a Design Engineer.
  • I greatly valued referring back to previous lectures and tutorials where specific skills were learnt which could be used during my placement.

Comments

No comments have been posted on this project yet.

Outdated Browser

This website has been built using some of the latest web technologies. Unfortunately, your browser doesn't support these technologies. To update your browser, please visit Outdated Browser.