Thesis Work - Reduced-Order Modeling and Transient Load Estimation in Nonlinear Driveline Systems

About this opportunity - Background

The main purpose of an automotive propulsion system is to deliver the required driving power to the wheels without experiencing critical failures. Additionally, to be competitive in the global market, the driveline must ensure high overall efficiency and drivability while minimizing noise and vibration. Designing a complex mechanical system requires detailed knowledge of its structural components and their interface connections, how they move and vibrate together during operation, and how they interact with distributed sensors and actuators.

Virtual simulation is a core tool in the development and analysis of complete automotive drivelines and their individual subsystems. Various physical aspects of the driveline are often simulated separately, using different system and subsystem model approximations that lack a clear relationship to one another or to a common, more general framework. As a result, when different models overlap or leave gaps in their prediction ranges, unwanted ambiguities are introduced into the overall analysis. This calls for improved analysis tools that enable accurate extraction of specific low-order models from a more general system representation, and their integration with physically measured response data.

Scope of the thesis work

Current efforts in mechanical design analysis aim to develop accurate low-order, low-frequency system dynamic models for estimation of distributed and propagated dimensioning loads that occur in Battery Electric Vehicle (BEV) drivelines. These models can be derived using advanced system identification techniques that combine measurement data with simulated sequential “snapshot” representations, extracted from existing high-fidelity dynamic models built on well-established physical principles. 

This thesis project will explore the application of system identification and model-order reduction methods to an existing nonlinear multi-body dynamic model. The specific case study involves a rotating system, such as an electric machine connected to a speed-reducing transmission with a differential gear and multi-link driveshafts. The goal is to extract low-order system model representations that maintain a known and traceable relationship to the parent model, and to use these representations for inverse estimation of transient drive torque corresponding to a previously measured rotational speed signal.

What you'll bring

We are seeking highly motivated students enrolled in a master’s program in Applied Mechanics, Systems Control, Mechatronics, or a related field. The ideal candidate is both analytical and practical, with an interest in numerical simulation and mathematical analysis of nonlinear dynamical systems. Experience, whether theoretical or practical, in multi-body dynamics, and structural dynamics is a bonus. You are expected to be working on-site.

Duration

•    The work will start in January 2026 
•    The duration for this thesis work is 20 weeks.
•    30 ECTS (Master academic credits) in agreement with your Thesis Advisor in University
•    This thesis is to be conducted by 1 Student or 2 Students working in pair.