Alia Salah introduces a multi-functional, model-based method for fault detection and identification in automotive electric machines. This approach integrates current vehicle diagnostics to detect faults early, before component failure. It utilizes digital twins and parameter estimation, alongside machine learning classification, to identify fault type and location. Moreover, it incorporates model reference adaptive control for fault-tolerant control, helping to maintain performance and ensure a safe driving experience.

Fault detection and identification utilizing digital twin and machine learning Leveraging the functionality of vehicle diagnostics and maintenance Fault-tolerant control for automotive electric machines

Autorentext

Alia Salah holds a doctoral degree in Automotive Mechatronics Engineering from University of Stuttgart, Germany. She is active in the electro-mobility field, automotive diagnostics and the development of control concepts for automotive electric machines. Her expertise extends to the development of anomaly detection concepts, predictive maintenance, and data science within the automotive domain. She has an extensive history of research publications in these fields.

Inhalt

Derivation of an equivalent mathematical model for the automotive electric power generation and storage system.- Providing a study on the effects of common faults on the performance of automotive electric machines.- Development of a concept for fault detection and identification for automotive electric machines.- Development of a concept for active fault-tolerant control based on model reference adaptive control.- Development of simulation models to prove the effectiveness of the proposed concepts.