To solve the fault diagnosis problem of the equivalent resistance change due to the degradation of electrolytic capacitor in the circuit, a fault diagnosis method based on inverse Kalman filtering is proposed. The converter is abstracted as a dynamic equation of Kalman filtering. It takes the parameters of circuit elements as unknown state of the Kalman filtering and uses the voltage and current of the circuit as the known matrix to inversely derive the recursion formula of Kalman filtering for parameter identification and fault diagnosis of converter. The parameter identification results for normal state of the converter show that the proposed inverse Kalman filtering parameter identification algorithm has higher accuracy. The fault diagnosis results for the circuit fault state show that the inverse Kalman filtering algorithm has good tracking ability, and it can detect faulty elements and estimate their parameters quickly. The simulations validate the effectiveness and practicality of the proposed method. This work is supported by National Natural Science Foundation of China (No. 61802150), Excellent Youth Foundation of Jiangsu Scientific Committee (No. BK20160001), Natural Science Foundation of Jiangsu Province (No. BK20170196), and Brand Specialty Construction Support Project of Jiangsu Province (No. PPZY2015A036).