Fault-Resilient Control Architecture for Electric Machine Drives in Mission-Critical Power Systems

Abstract

In power systems where fault tolerance is critical to the continued uptime of a system, and operational reliability is critical to system safety, electric machine drives have a key role. Nonetheless, the traditional control methods tend to reduce their performance in fault conditions because they support lesser adaptability and slower compensation in fault situations. It is suggested in this paper that a new fault-resilient control architecture should be developed, combining the real-time fault detection, estimation and control reconfiguration into a single architecture. An electric drive state-space model that is dynamic is developed and fault injection mechanisms are included to model sensor and actuator anomalies. A fault detection and estimation scheme is implemented using observers and is used to ensure that system deviations are correctly determined whereas an adaptive control law replenishes the effect of faults to keep the system stable and performing. Lyapunov-based analysis is applied so as to ensure that the system stability is guaranteed in different fault conditions. The suggested methodology is confirmed by simulation experiments in different fault conditions, such as sensor bias, inverter faults, and load disturbances. Compared to conventional and robust control methods, comparative study results show that speed tracking accuracy is greatly improved, torque ripple is also reduced and also recovery time is found to improve rapidly. The results determine that the proposed architecture is effective in improving the resilience and reliability of electric machine drives in mission critical applications.