Research And Implementation Of Analog Circuit Diagnosis System Based On BP Neural Network And DSP Technology

The applications of large-scale integrated circuits allow the scale and the structure of a network to be more and more functioned and modularized with the development of modern electronic industry. So it is an urgent subject in practical project to study how to identify the fault elements correctly by using modern diagnosis technique for large-scale circuits with tolerance, and it is also a key step for practical application of fault diagnosis for analog networks theory and technology.This dissertation based on NNs profoundly studies fault feature extraction and fault diagnosis methodology. It academic foundation is modern test technology,signal processing , information fusion and testability analysis,etc. the analog circuit soft fault diagnosis is realized to select test electric circuit by using TMS320F2812 under the Components existence tolerance condition . Author's mian work concentrates on eight aspects as follows:1,The application of BP neural network in analog circuit faults diagnosis and the fault features extraction are discussed in detail . The training program for weight values and threshold values is compiled under Matlab environment, and the simulation results validate the feasibility of the method.2,The simulation source program of the measuring electric circuit is compiled under PSpice4.02 environment, and Montecarlo analysis is carried on to obtain the neural network trained data sample set.3,The fault sample set is established by using the multi-frequency combine method.4,The four kinds of frequencies PWM signals are generated by using DSP, which are passed through four groups three steps filer circuit to obtain needed four kinds of frequencies excitation signals.5,By using the TMS320F2812 performance characteristic , the corresponding interface circuit is designed, and its PCB board is made.6,The digital filter program is compiled under MATLAB Environment, and the filter Coefficient is obtained.7,Under the CCS integrated development environment, the corresponding AD data acquisition programs,the SCI communication programs,the PWM Waveform programs and the LCD realization programs are compiled. The PC receive programs are compiled under VB environment.8,The result that the actual tolerance circuit is tested indicates the diagnosis system is of useful value.