A Fault Prediction and Health Management Method Applied to Circuits and Systems

Abstract

The invention belongs to the technical field of circuits and systems, and provides a deep learning-based fault prediction and health management method applied to circuits and systems, which is used for real-time monitoring of circuits and system equipment in operation; the method first prepares multiple waiting Test the circuit, carry out the aging test separately, obtain the training data and store it in the database; then conduct PCA analysis to obtain the training sample; then use the deep learning model to train the neural network and put it into the test chip; finally use the test chip to check the working status The health status of the test circuit is monitored in real time, and its remaining life is calculated at the same time. The invention can monitor the operating status of circuits and system equipment in real time, predict the time of failure, reduce the probability of sudden failure, and avoid many potential safety hazards when sudden failure occurs, thereby protecting property and reducing maintenance expenses.

Description

technical field [0001] The invention belongs to the technical field of circuits and systems, and in particular relates to a fault prediction and health management method based on deep learning applied to circuits and systems. Background technique [0002] With the development of science and technology, the integration and complexity of modern equipment are increasing day by day, and the maintenance cost and difficulty are also rising sharply; the traditional manual maintenance method with low efficiency and high cost is facing a large number of complex modern equipment time is no longer applicable. [0003] Failure prediction and health management (PHM, Prognostic and Health Management) technology aims to reduce the cost of manual maintenance, to analyze the failure model of various equipment to determine its health status, so as to perform self-assessment and failure warning in an unattended situation. [0004] PHM technology generally has the capabilities of fault detecti...

AI Technical Summary

Problems solved by technology

"The method described in this paper is to predict the life of the target by establishing a physical model. The disadvantage of this method is that the physical model used to predict the life requires a relevant professional background, and its adaptability is narrow. The accuracy will be greatly reduced in the presence of the situation, so it cannot be put into use in time, and additional manpower will be required

Images