Application of extension neural network algorithm and chaos synchronization detection method to partial discharge diagnosis of power capacitor

•The proposed method (ENN & Chaos) can reduce huge partial discharge data greatly.•Any subtle changes in capacitor discharge signal can be detected effectively.•Results show that ENN has the best recognition rate compared to other algorithms.•The self-developed human-machine interface is integrated with chaos and ENN.•The study may serve as a basis for fault diagnosis in the field of capacitors.