Hunan University launches new diagnostic theory and method for analog circuits and switched capacitor networks

fighting

Hunan University launches new diagnostic theory and method for analog circuits and switched capacitor networks [Copy link]

Hunan University launches new diagnostic theory and method for analog circuits and switched capacitor networks

Luo Zhongyou Huang Hongli

This newspaper reports that　more than ten experts including Professor He Yigang of Hunan University, Professor Luo Xianjue and Professor Qiu Ziyuan of Xi'an Jiaotong University have worked together for nearly nine years and have recently launched a new set of fault diagnosis theories and methods for analog circuits and switched capacitor networks, funded by five scientific funds including the National Natural Science Foundation and the University Doctoral Program Fund. This achievement has well solved the problem of fault diagnosis caused by the continuity, tolerance and wide range of component parameters. The expert appraisal organized by the Hunan Provincial Department of Science and Technology recently concluded that the overall achievement has reached the international advanced level, and is at the international leading level in terms of analog integrated circuit neural network methods and switched capacitor network fault diagnosis. 　　Exploring new theories and methods for fault diagnosis of analog circuits and switched capacitor networks is a major research topic at the forefront of international circuit theory. At present, the traditional theories and methods based on von Neumann computers are widely used internationally. This method has a low accuracy rate (about 60%) and can often only measure hard faults. The diagnostic theories and methods proposed by experts from Hunan University and Xi'an Jiaotong University are based on pattern recognition and neural networks, with a misdiagnosis rate of less than 1%. They are highly applicable, fast in diagnosis, and have high positioning accuracy. In this way, component faults and parameter faults can be measured in real time during the production stage of large-scale integrated circuits and printed circuit boards (PCBs), and it is as convenient as looking up a Chinese dictionary, which can greatly improve the yield rate. Identification experts believe that the switched capacitor network fault diagnosis theory and method fills the research gap in this field at home and abroad, and will have a great role in promoting the development of my country's microelectronics technology, analog integrated circuits and chip systems.