Adaptive Control applied to thermal management of electrified heavy ve

Sustainability including climate change are the challenges of our generation. Our contribution is to offer leading transport and infrastructure solutions enabling societies to prosper in a sustainable way. At Volvo group, we are committed to the ambitions and climate change goals of the Paris Agreement. From a lifecycle perspective, most of the emissions occur during the use phase of our products. Therefore, our priority is to develop solutions that reduce the carbon emissions from transportation. Today Volvo is the market leader in offering full range of electric trucks.
We are Vehicle Energy Management team at Group Trucks Technology (GTT). As the team’s name suggests, we work in the area of energy management, where we optimize the vehicle energy using the predicted information such as topography, curvature, and speed limits from the road ahead of the vehicle.
We have been working in this technology area for several years with several successful thesis projects. This thesis work will have as a starting point a master thesis work carried out during 2023 on “Optimal predictive control applied to thermal management of electrified heavy vehicles “[1]. What we expect from this master-thesis project is to build upon the knowledge we have gained so far and apply it to our existing Battery Electric Vehicles (BEV) and later on to Fuel Cell Electric Vehicles (FCEV).


Thesis background


The thermal system has gained in importance when the industry transit from combustion engines (ICE) to BEVs. The reason is that good driving range and low cost are key properties of a successful introduction of BEVs on the market, and controlling the thermal system is an important enabler to improve range and reduce costs for electrified heavy-duty trucks.
The purpose of the controller serves several, and sometimes contradictive, purposes. On one hand the controller aims to minimize the energy consumption of the cooling system actuators, but the purpose is also to make sure that the battery doesn’t lose its capacity over time. In addition, it is also important to make sure that the motors and the battery can deliver the power requested by the driver and is not limited due to high temperatures in the motor and battery.


To summarize, predictive optimal control of the thermal system is an important technology for bringing electrified heavy-duty vehicles to the market.
Problem motivating the thesis


Keywords: thermal management, energy optimization, predictive strategies, MPC, Adaptive Control, Machine Learning, BEV, FCEV


Model uncertainty is the biggest challenge when applying model-based predictive optimal control to the thermal system. On one hand, the model needs to be sufficiently accurate to use for prediction and optimal control, but it should be sufficiently simple to use in an embedded controller. And even if the model is accurate in a simulation environment, the controller still needs to be adjusted when it is used in a real application.
In the prior thesis work, Liljeqvist [1] developed models designed for optimal control and applied it to a receding horizon controller. In this thesis work you will further explore this topic by focusing on robustness to model errors. In particular, you will investigate estimation techniques that are suitable for adaptive non-linear control. It could be classical techniques such as recursive least squares, linear parameter estimation or more recent techniques developed in the area of machine learning. Another approach to achieve robustness that could be considered is to introduce feedback control to the output of the predictive controller.
Previous Master theses in our team
[1] Liljeqvist, Max. Optimal predictive control applied to thermal management of electrified heavy vehicles, 2023. Available: online.


Objective or Research Question
What parameters/states should be monitored/estimated in an adaptive fashion to improve robustness to model uncertainty?

What method should be used to estimate uncertain parameters?
How can feedback control be used together with the predictive controller to improve robustness?

Deliverables (flexible)
Evaluation of different estimating techniques
Implementation in MATLAB /Simulink
Evaluate improved robustness to model errors

Requirements on student background
Master students in Automotive, Control, Mechatronics or Engineering Physics
Interest in Control, Optimization
Good knowledge of MATLAB/Simulink
CV and motivation letter required
Supervision and examination
Safe and Efficient Driving, GTT, Volvo Group
Chalmers University of Technology

Thesis Level: Master
Language: English
Starting date: Spring 2024
Number of students: 1
Physical location: Mostly at Volvo Lundby.
Examiner: Torsten Wik


Volvo contact persons: Olof Lindgärde (olof.lindgarde@volvo.com)


Kindly note that due to GDPR, we will not accept applications via mail. Please use our career site.