Direkt zum Inhalt springen
login.png Login join.png Register    |
de | en
MyTUM-Portal
Technical University of Munich

Technical University of Munich

Sitemap > Bulletin Board > Diplomarbeiten, Bachelor- und Masterarbeiten > MS thesis: Predictive energy management using out-car information
up   Back to  News Board    previous   Browse in News  next    

MS thesis: Predictive energy management using out-car information

20.07.2021, Diplomarbeiten, Bachelor- und Masterarbeiten

The range of electric vehicles depends significantly on optimal heating or cooling of batteries as well as prediction of energy demand. In this thesis, we aim to optimize the vehicle range and energy efficiency by predicting the expected trip, also based on external context information from data market places. /p>


The transport sector is facing major social challenges: traffic growth, demographic change, ensuring affordable mobility, e-ticketing for all areas of use, increasing requirements for energy efficiency and the reduction of CO2 emissions.

Battery Electric Vehicles (BEV) can adjust the battery’s thermal management to optimum conditions, which can reduce energy consumption significantly. This works best when the route is actively guided by the driver, and a charge stop is planned. One challenge arises because few people enter every destination into their navigation system. Without this information the system can only be reactive. Other challenges relate to changing environmental conditions leading to excessive energy usage, i.e. traffic and weather.

State of the art BEV’s are using in-car prediction algorithms to predict the upcoming course of the journey. This look into the future prevents unnecessary heating or cooling, which saves energy and increases the range, however due to the lack of (correct-) input data the future time horizon is limited, and a non-global optimum is reached. To further optimize the vehicle’s energy usage more input and more data is required. Driving behaviour, actual traffic data, local weather information, charging station availability and routing information are promising data sources to increase the vehicle’s prediction accuracy and further optimise the energy usage, however so far only limited services are available that possess this kind of information in a secure and trusted way.

Recently, new technology for trusted data exchange has been developed and data market places become available. The main technologies are “Data Spaces” and the main drivers are the European Gaia-X initiative.

Purpose:

The purpose of this study is to integrate and evaluate a service that handles several types of prediction functionalities by the usage of available cloud data. This service can be called by individual vehicles to improve their energy usage and save CO2 and increase driver’s satisfaction. This study would include a proof of concept of using this secure trusted data exchange that benefits the battery electric vehicle energy consumption.

This Master thesis project consists out of the following tasks:

• Data study: To create an overview and understanding of available cloud information which are currently available

• Problem definition: Define specific use cases for the thesis and a solution approach

• Concept definition: Design and implement a predictive energy system in a prototype

• Concept evaluation: Evaluate the system, in particular to the use cases

Project Targets:

• Literature study

• Data handling

• Prediction features in cloud

• Optimization features in cloud

• Control study of vehicle model

• Simulations using cloud emulator

• Data analysis

• Reporting

Requirements:

• Currently enrolled in a computer science / engineering Master studies with a focus on data analysis, automotive, mechanical engineering or similar

• Good English skills both written and spoken

• Has an interest to work in the automotive area

• Programming skills are required (e.g. Java, Python, MATLAB and Simulink)

• Good data analysis skills

• Good MS-Office skills are required (especially Word, Excel, PowerPoint)

Kontakt: prehofer@in.tum.de

Todays events

no events today.

Calendar of events