Application of DBN and GWO-SVM in analog circuit fault diagnosis

Abstract

For large-scale integrated electronic equipment, the complex operating mechanisms make fault detection very difficult. Therefore, it is important to accurately identify analog circuit faults in a timely manner. To overcome this problem, this paper proposes a novel fault diagnosis method based on the deep belief network (DBN) and restricted Boltzmann machine (RBM) optimized by the gray wolf optimization (GWO) algorithm. First, DBN is used to extract the deep features of the analog circuit output signal. Then, GWO is used to optimize the penalty factor c and kernel parameter g of support vector machine (SVM). Finally, GWO-SVM is used to diagnose the signal features extracted by the DBN. Fault diagnosis simulation was conducted for the Sallen–Key band-pass filter and a four-opamp biquad highpass filter. The experimental results show that compared with the existing algorithms, the algorithm proposed in this paper improves the accuracy of Sallen–Key bandpass filter circuit to 100% and shortens the fault diagnosis time by about 90%; for four-opamp biquad highpass filter, the accuracy rate has increased to 99.68%, and the fault diagnosis time has been shortened by approximately 75%, and reduce hundreds of iterations. Moreover, the experimental results reveal that the proposed fault diagnosis method greatly improves the accuracy of analog circuit fault diagnosis, which solves a major problem in analog circuitry and has great significance for the future development of relevant applications.

Improved Random Forest Algorithm Based on Decision Paths for Fault Diagnosis of Chemical Process with Incomplete Data

“…BP, which was developed in 1986, by scientists led by Rumelhart and McClelland, is a multi-layer feed-forward neural network that is trained by error reverse propagation algorithms and is the most widely used neural network. DBN consists of several restricted Boltzmann machine (RBM) layers and a BP layer, and its unsupervised pre-training can extract high-level abstract representations from the input data [58][59][60]. RBF is often used to approximate multi-dimensional surfaces by a linear combination of radial symmetric functions, based on Euclidean distance.…”

Research on fault diagnosis method of aviation cable based on improved Adaboost

“…In recent years, many scholars at home and abroad have also proposed methods that are different from traditional cable fault detection to adapt to the current complex aircraft cable fault detection. Foreign scholars such as Su et al 10 proposed a Support Vector Machine (SVM) fault diagnosis method, which can accurately classify fault targets and effectively improve the accuracy of fault classification; Dhumale et al 11 proposed a fault diagnosis method based on Artificial Neural Network (ANN), which trains the artificial neural network by collecting the characteristic data of the fault, and uses the trained artificial neural network to analyze the existing fault diagnosis is simpler than before, and the diagnosis result is more efficient; Han et al 12 proposed a fault diagnosis method based on Decision Tree, which can identify the type of fault more accurately, intelligently, intuitively, and efficiently, and greatly shorten the time of fault diagnosis. Domestic scholars such as Wu et al 13 proposed a cable fault detection algorithm based on Spread Spectrum Time Domain Reflection (SSTDR), which uses a direct sequence spread spectrum signal as a detection signal.…”

TLSCA-SVM Fault Diagnosis Optimization Method Based on Transfer Learning

“…Secondly, the fault data is processed, using methods such as feature-value extraction, dimensionality reduction, linearization, etc. Finally, the processed data is used in the modelling of fault diagnosis to judge the performance of the model [14]. In this article, the data of Sallen-Key circuit is used for model comparison and the more complex CSTV circuit is used for validation to determine the versatility of the model.…”