The Research Of The Analog Circuit Fault Diagnosis Method Based On Wavelet Entropy Transform And Adaptive Quantum Particle Swarm Optimization Algorithm

As modern electronic technology develops very rapidly, the research in the fault diagnosis theories and methods for analog circuits is becoming not only very popular but also challenging. The modern electronic technology of development asks for higher demands of testing and the fault diagnosis on analog circuit a higher demand, therefore, traditional fault diagnosis theories and methods cannot meet actual the requirement due to difficulties inherent in analog circuits fault diagnosis. On the other hand, computation intelligence technologies including neural network method, evolutionary computation and something else are of great interest to researchers, which might provide Potential solution to fault diagnosis.The paper analyse research presented situation of the analog circuit fault diagnosis method, discuss the difficulty and classification method based on the analog circuit fault diagnosis, Then analog circuits fault diagnosis theory is researched deeply.Because analog circuits are usually with tolerance and the voltage and current of different nodes are sensitive to different fault components, the Montecarlo analysis and the worst case analysis special functions of PSpice simulated treatment diagnosis circuit. Based on these research, the paper does some research in the information entropy, the maximum energy method principle, the wavelet transform method and the wavelet entropy method. This thesis bases on information theory and the wavelet transform, proposing wavelet entropy method, using the wavelet transform method and the wavelet entropy separately extracted fault feature on treated diagnoses circuit.The radial basis function neural network is a novel effective forward feed neural network; it has optimal approximation and the overall situation most superior performance. Researched the intelligent optimization algorithm by the system, combined the quantum particle swarm optimization and the adaptive quantum particle swarm optimization with neural network algorithm, proposes the analog circuit fault diagnosis method based on the radial basis function neural network and the adaptive quantum particle swarm optimization algorithm. The RBF neural network with QPSO and AQPSO is trainedBased on Matlab the R2008b platform, fault diagnosis system of analog circuits is developed, several modules of fault diagnosis system are designed, realized the analog circuit fault diagnosis method which combined the radial direction primary function neural network with the adaptive quantum particle swarm optimization algorithm. The paper has done the massive simulation experiment, to the paper discussed different neural network diagnosis method, the wavelet transformation and the wavelet entropy two feature extraction methods has carried on the comparison. The experimental results indicated that this article gives based on QPSO RBF and the AQPSORBF analogous circuit failure diagnosis method superiority.