Design Of A Components Degradation Parameter Test System And Establishment Of Degradation Model

As smart devices are often in a long working state,the parameters of the components inside the electronic system under a long period of powering up will change,which will lead to changes in their performance so as to affect the performance of the entire electronic system and even the entire smart device.Therefore,this topic focuses on the performance changes of typical components in commonly used control system circuits under long-term power-up conditions.The research is of great significance for understanding the fault evolution of the whole electronic system and enhancing the system reliability.In this paper,the typical typical degradation-prone devices in the electronic system: thin film resistors,electrolytic capacitors,opto-couplers,electromagnetic relays,and the like are studied.MOSFETs as well as operational amplifiers are analyzed for degradation mechanisms and their degradation parameters are clarified.Based on the clarification of the degradation parameters and the corresponding technical specifications,the hybrid bus of LAN and USB is used as the basis,together with the The Labwindows/CVI-based design test software is combined with an automatic measurement system for device parameters.Based on this automatic measurement system,this paper has completed the measurement and acquisition of degradation parameters of each component for 480 hours,and has also measured the degradation parameters of the measurement circuit.The resulting measurement errors are theoretically analyzed and corrected.Finally,the grey model,support vector machine regression,and generalized neural network,which are three commonly used smallsample modeling algorithms,are used to analyze the measurement errors of each device.The trend of evolution of degradation parameters is modeled,and after deriving the model and comparing it with the existing literature and the analysis of the device's technical manuals,this article The resulting degradation parameter model for each device can characterize the performance parameters of each device under the corresponding operating conditions,and can also be used in the More accurate prediction of parameter changes over time,thus providing an aid to the analysis of fault evolution patterns in the entire electronic system and providing The predictive maintenance of the system provides theoretical support.