Research And Implementation Of Analog Circuit Fault Diagnosis Based On Neural Network And Reserve Pool Calculation

So far,the proportion of the analog circuit is still high in the electronic equipment.With the popularity of electronic equipment,the application of analog circuit is more and more widely,which is of great importance to analog circuit fault diagnosis and prevention.As a non-linear information processing technology,neural network is easy to implement,which has a strong self-learning and adaptive ability.The reserve pool technology is through the internal structure of the reserve pool,mutual coupling nonlinear node to simulate the activities of the neurons in the brain,the reserve pool can be done in a very short period of time the complex signal identification,classification and prediction functions.This thesis makes a deep theoretical simulation research of the fault diagnosis method of the analog circuit based on the traditional neural network and the single chaotic node reserve pool computing.(1)Extracting the fault characteristic information of analog circuit using principal component analysis and wavelet transform method.The original high-dimensional data reduced to two dimensions by using principal component analysis.Time-frequency analysis of original data and extracting the fault feature using the method of wavelet transform.(2)Researching analog circuit fault diagnosis based on BP neural network,analyzing the ac amplitude frequency characteristics of Sallen-Key band-pass filter and studying components which make large effects on circuit performance.(3)Put forward the analog circuit fault diagnosis new method based on single chaotic node reserve pool computing,analyzing the structure and principle of reserve pool computing and researching on the method of chaotic signal based on Ikeda delay feedback system.And calculation of reserves pool in analog circuit fault diagnosis feasibility is analyzed.Through the above the main work,the standard BP neural network algorithm and the LM optimization method for fault diagnosis,combined with the convergence steps and diagnostic accuracy,LM optimization method is the best.On the basis of the traditional neural networks algorithm,based on a single chaotic node calculate reserve pool of the analog circuit fault diagnosis method of the diagnostic accuracy of the new and the diagnostic accuracy of BP algorithm is close to,therefore,based on a single chaotic node calculate reserve pool of new type of analog circuit fault diagnosis method is feasible.