Research On Intelligent Fault Diagnosis Technology Of Analog Circuits

With the development of electronic industry, the integration of analog circuits is increasing. Requirements on technology of analog circuits fault diagnosis become higher than before. Because of the inherent characters such as the continuity and non-linearity of responsiveness in analog circuits, expected effects are difficult to achieve in practical application of traditional manual testing and fault diagnosis. Based on the theory of Neural Network, the method of fault feature extraction and fault diagnosis were intensively researched in the paper, and a intelligent fault diagnosis system was designed based on the technology of virtual instrument.Based on the fault types of analog circuits and fault diagnosis of Neural Network, a fault diagnosis method of analog circuits based on BP Neural Network was proposed. Contrary to the long training time of BP Neural Network, two improving methods of BP Neural Network were explained: additional momentum method and LM algorithm. The simulation model of analog circuits was established in the Multisim platform, and the equidistant sampling method was used to extract fault feature of the output responsiveness, the feasibility of additional momentum improved BP Neural Network was proved. In order to get the fault feature of output effectively, the Wavelet combined with Neural Network method for analog circuits fault diagnosis was researched in the paper. Using Wavelet decomposition to process the impulse response drastically reduce the number of input fed to the Neural Network improved by LM algorithm. It was proved that the correct rates of those two methods were high, in contrast, the Wavelet combined with Neural Network was better in correct rate and speed.Finally, an intelligent fault diagnosis system was designed in LabVIEW. The front panel displayed fault types, and the Matlab script nodes connected the LabVIEW and Matlab. When running the LabVIEW program, the Matlab program can automatically be called to training and testing. Application of virtual instrument to the analog circuits fault diagnosis system, have a good engineering application prospect.