Abstract
Traditional analog circuit fault diagnosis usually consists of feature extraction, feature selection and fault classifier. However, fault feature extraction and selection take much manual effort, which make the design of the fault diagnosis system complicated and not universal. This paper proposes an analog circuit fault diagnosis method based on Extreme Learning Machine (ELM). The proposed method uses the single pulse as the stimulus signal to circuit-under-test (CUT), and then the raw time domain response of CUT is directly sampled to construct fault samples, which are inputted into the ELM network to obtain fault diagnosis results. The experimental results show that the average diagnostic accuracies reach 100% and 99.5% on the Sallen-key band-pass filter circuit and the four opamp biquad high-pass filter circuit, respectively. The algorithm achieves good diagnostic accuracy and efficiency in the absence of feature extraction and selection.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call