Model-Based Safety Analysis of Power Electronic Control Using AUTOSAR Safe State Management

Authors:
Siddhesh Pimpale

Addresses:
Department of Electric Vehicle, Dana Inc., Novi, Michigan, United States of America.

Abstract:

Functional safety in power electronic control systems is an important requirement, particularly in the safety-critical application areas such as automotive, aerospace, and industrial automation. This paper proposes a model-based safety analysis for power electronic control by means of AUTOSAR Safe State Management. Automotive Open System Architecture (AUTOSAR) is a standards-based approach for developing and integrating Electronic Control Units (ECUs) to automate the electronic systems in automobiles. In this paper, model-based systems engineering (MBSE) is employed to model, simulate, and validate the behaviour of the power electronic controller under fault conditions. This would enable plant safety objectives and Technical Safety Requirements (TSRs) to be incorporated into the system model, helping to provide early hazard propagation path analysis and ensuring Ethical Compliance with ISO 26262 regulations. Simulation scenarios are used to demonstrate how the system transitions to safe conditions when overcurrent, short-circuit, and control loop failures occur. Analysis results show that model-based application of AUTOSAR Safe State Management increases traceability, reduces application development effort, and enhances the overall safety integrity of the system. This study demonstrates the significance of integrating the safety aspects of AUTOSAR with MBSE tools to model a reliable/fault-tolerant control system for power electronics.

Keywords: Functional Safety; Power Electronics; Systems Engineering; Hazard Propagation; Simulation Scenarios; Safe State Management; Safety Integrity; Electronic Control Units.

Received on: 13/07/2024, Revised on: 15/09/2024, Accepted on: 12/10/2024, Published on: 09/03/2025

DOI: 10.69888/FTSESS.2025.000371

FMDB Transactions on Sustainable Environmental Sciences, 2025 Vol. 2 No. 1, Pages: 15-26

  • Views : 44
  • Downloads : 7
Download PDF