Analog Circuit Fault Diagnosis Methods Based On Chaotic Neural Network

With the advances in electronic technology, the analog circuits develop toward the large-scale, integrated direction. The probability of components failure is greatly increase, at the same time, failure of individual components will affect the whole system. The developments of technology put forward strict requirement to analog circuit fault testing and diagnosis, so the research of analog circuits fault diagnosis is particularly important. Some of the traditional fault diagnosis methods are no longer applicable to the present circuits. With the artificial intelligence technology such as neural network and wavelet decomposition, chaos, particle swarm optimization algorithm are applied to the fault diagnosis, open up a new way for analog circuits fault diagnosis. Research on simple, efficient methods is development trends of analog circuit fault diagnosis. This paper discusses the process of analog circuit fault diagnosis and studied the analog circuit fault diagnosis method based on neural network and chaotic neural network. The main contents include following parts:1. The article elaborate the theory of neural networks, study the flexible application and effective classification performance of neural networks and use wavelet neural network for analog circuits fault diagnosis.2. In the session of fault feature extraction, this paper propose a method of extracting fault features based on wavelet decomposition and fuzzy clustering which can effectively extract the fault feature to reduce the dimensions of the input of neural network.3. A method of analog circuit fault diagnosis based on chaotic neural network is proposed in this paper. The approach are combined the chaotic motion’s ergodic, randomness, sensitive to the initial value with the mapping and learning ability of neural network which make the network have the chaos characteristics and make the neural network get a better convergence, learning and generalization ability. It can improve the efficiency of fault diagnosis by using the chaotic neural network.4. In order to verify the correctness of the proposed method in this paper,use MATLAB and ORCAD simulation software to do the actual circuit simulation. The simulation result show that the proposed method can improve the efficiency of analog circuit fault diagnosis and it has strong capacity of soft fault detection.