Analog circuit fault diagnosis based on enhanced <scp>Harris Hawks</scp> optimization algorithm with <scp>RBF</scp> neutral network

Circuit faults are caused by the change of device parameters in analog circuit. Aiming at problems that fault feature extraction is difficult and signal cannot be effectively classified, an enhanced Harris Hawks algorithm proposed to optimize parameter optimization process RBF neural network, so as realize identification diagnosis Based on wavelet packet analysis, output response circuit decomposed, vector extracted. Taking power conversion electronic interlocking system research object, 500 sets data collected, EHHO-RBF trained tested different faults, compared with other network algorithms, experimental results show accuracy method about 96.5%, which verifies effectiveness feasibility algorithm.