Fault mitigation in an interconnected network using artificial intelligence technique

The reliability of an interconnected power network is critical for maintaining a stable and efficient power supply. This research focuses on the detection, classification and mitigation of faults on the Ibom power interconnected network comprising of three synchronous generators, assuming each generator generate a 40 MW power, with a total power capacity of 120 MW power after synchronization. The network transmits it power through four transmission stations, first Afaha-Ube sub-transmission station, which has a load offtake of 60 MW, Itu sub-transmission station with 20 MW, Eket sub-transmission station with 25 MW, and Ekim sub-transmission station with 15 MW. All the sub-transmission station received 132 kV and step it down to 33kV. The system was first analyzed under normal operating conditions before various types of faults were introduced at the Afaha-Ube Substation bus, these included single line-to-ground (A-G), double line-to-ground (AB-G), and three-phase-to-ground (ABC-G), faults. To enhance fault detection, classification and system protection, current signals only were obtained through discrete wavelet transform (DWT). These signals were fed as input to train the ANFIS to properly detect and classified each fault type The ANFIS model was also trained to analyze fault current patterns and trigger a trip signal to the circuit breaker for fault isolation. Although faults persisted from 0.8 to 1.6 seconds, ANFIS was optimized to identify the fault and send a trip signal at 1.2 seconds, effectively mitigating damage and improving network stability. The result obtained shows that ANFIS successfully trip the Afaha-Ube bus station at exactly 1.2 seconds.