Decentralised fault detection, identification, and mitigation in excitation control systems

• Novel fault detection, identification and mitigation method for excitation system.
• This method is capable of detecting excitation system faults in near real time.
• Proposed method is capable of restoring the system to its pre-fault condition.

A new intelligent, decentralised, fault detection, location identification and mitigation approach for the excitation control systems of multi-generator power grids is presented in this paper. The proposed approach detects and locates faults in basic components of the excitation systems with a high accuracy and in almost real-time using interval type 2 fuzzy set theory. Once a fault is detected and the location of the fault is identified, a mitigation action follows, where the signal out of the faulty component is automatically replaced by a stabilising signal generated by the fault mitigation system. The mitigation action ensures that the field excitation voltage (thus the terminal voltage of the affected generator) is controlled such that the affected generator and the whole system maintain their pre-fault operating condition. The high accuracy and fast response time aspects of the proposed approach are verified through case studies performed on the IEEE 30 bus test system.