The Design And Implementation Of DSP-based Fault Diagnosis For The Analog Circuit Using Neural Network

With the rapid development of the Electronic Industry, the importance of the analog circuit fault diagnosis is more and more obvious, it has important signification for working orderly and dependability design of electronic equipment or system. On the base of traditional diagnosis and theory, this paper studies the application of neural network and DSP in the analog circuit fault diagnosis.In the first part of this paper, some algorithms usually used are introduced,Then,the application of BP neural network to the analog circuit fault diagnosis system is discussed especially, which has the robustness, associated memory and nonlinear mapping, this paper embodies ingathering of samples, confirmation of framework and training of network. Improve samples with circuit emulator OrCAD/PSpice9.1 to improve recognition ability of network, and a program is designed for training the BP neural network by MATLAB. In the second part, We have used the DSP chip of TMS320F2812 as the controller of build-in system based on the characteristics of analog circuit. This chip is a new product released by TI. Its speed of instruction performance is 150 MIPS, which is nearly one hundred times of other MCU's.TMS320F2812 has been built in high-speed A/D conversion cell and other external devices. We have studied the characteristics, capabilities, parameters and external circuit demands of this chip deeply, have completed circuit design and algorithm program. At last, an actual circuit is selected to test the system. The result of experiments indicated that our system works steady and performs well.Practice has proved that the application of BP neural network and DSP mentioned above are all effective, they have great applying value and developing potential in analog circuit fault diagnosis.