A new fault diagnosis approach for analog circuits based on spectrum image and feature weighted kernel Fisher discriminant analysis.

Analog circuits are one of the most commonly used components in industrial equipment and, therefore, circuit failure may lead to significant causalities and even huge financial losses. To address this problem, this work presents a fault diagnosis method based on spectrum images for analog circuits. Unlike traditional analysis methods in a one-dimensional space, this study employs a computing method in the field of image processing to realize automatic feature extraction and fault diagnosis in a two-dimensional space. The proposed approach mainly involves the following steps. First, the sampling signals are converted into spectrum images by utilizing cross-wavelet transform, which can be further processed by the following image-based feature extraction method. Then, Krawtchouk moment is applied to extract both the global and local features of the spectrum images and finally form the feature vector. Feature weighted kernel Fisher discriminant analysis is then introduced for locating faults. Two typical analog circuits, video amplifier circuit and opamp high-pass filter circuit, are chosen to demonstrate the effectiveness of the proposed method. Simulation results show that the proposed approach based on spectrum images achieves a high accuracy, thus providing a highly effective means to fault diagnosis for analog circuits.