Creating a simulation model to verify efficient engine shutdown using map data

No Thumbnail Available

URL

Journal Title

Journal ISSN

Volume Title

School of Electrical Engineering | Master's thesis

Date

2024-11-17

Department

Major/Subject

Autonomous Systems

Mcode

Degree programme

Master's Programme in ICT Innovation

Language

en

Pages

50

Series

Abstract

As of 2023, approximately 1.45 billion vehicles operate globally, contributing to around 24% of CO2 emissions. Frequent starting and stopping of vehicle engines exacerbates emissions. Although some studies apply engine start-stop technology to reduce emissions, they primarily target private vehicles, with limited application in public transport systems. Moreover, the logic used in these applications often results in frequent start-stop actions, which consequently reduces the lifespan of the starter motor. In this research, we develop machine learning models specifically tailored to the stop-and-go patterns of public buses. We use public transit map data to train and evaluate three models—Random Forest, XGBoost, and Keras—based on their dwell time prediction performance. We then select XGBoost for its accuracy and integrate its predictions into Scania's UF568 start-stop system, enabling data-driven engine optimization, while also developing our own Electronic Control Unit(ECU) application. Our work improves fuel efficiency and supports the extended lifespan of bus starters, contributing to a more sustainable public transportation system.

Description

Supervisor

Baumann, Dominik

Thesis advisor

Tota, Pranay
Berlin, Filip

Keywords

engine start-stop system, map data, machine learning, starter lifespan optimization, fuel efficiency, ECU

Other note

Citation