Font Size: a A A

Research On Life Prediction Of Analog Circuit Based On Physical Failure-electrical Characteristic Analysis

Posted on:2019-01-04Degree:MasterType:Thesis
Country:ChinaCandidate:Q WangFull Text:PDF
GTID:2348330563954023Subject:Instrument Science and Technology
Abstract/Summary:PDF Full Text Request
Analog circuit play a very important role in weapons and equipment,meanwhile,a small problem in analog circuit may cause a military accident.In order to reduce the failure rate of analog circuit,accurate prediction and timely maintenance are very necessary.At present,a large number of studies on the reliability of analog circuit have been conducted at home and abroad,which are mainly divided into data-driven and physical analysis methods.Today,due to the prevalence of big data,data-driven methods have achieved very good results.However,the complexity and diversity of circuit structures are difficult to overcome,leading to slow development of physical analysis methods.For some military equipment,due to the limitations of the device's own structure and usage scenarios,it is difficult to obtain performance data at different time stages.So,only physical analysis of its internal structure can be used to predict equipment life.Therefore,this paper takes analog circuit life prediction based on the relationship physical failure and electrical characteristic as the research topic.The main research contents are divided into the following four parts:1.Component electrical characteristics parameter degradation modeling based on physical failure analysis.Firstly,the degradation model of resistors and capacitors is analyzed.Then the degradation effect of MOSFET devices is studied.It is shown that the main cause of MOSFET device degradation is due to the occurrence of hot carrier effect and negative bias temperature instability inside the device.Then the degradation model of MOSFET device is established by analyzing its generation mechanism.Finally,the degradation model was verified with Silvaco TCAD software.The results show that the results of degradation model is basically consistent with the simulation results.2.Degradation trend prediction method based on sensitivity analysis of analog circuit: this paper studies an algorithm for the sensitivity analysis of the circuit.Firstly,the sensitivity of the component to the circuit output is calculated based on the transfer function of the circuit,and then the relationship between the component parameters and the output of the circuit is established.Last,the degradation of the circuit performance is analyzed when the device parameters are degraded.3.Research on circuit degradation behavior.According to the degradation model of the components,the degradation behavior of the circuit is analyzed.First analyzing the behavior of the circuit without considering degradation,and establishing its behavior model.Then,the relationship between the circuit input and the component degradation model parameters is analyzed after the introduction of the component degradation model,and the component input is converted into the circuit input in the component degradation model,so that the relationship between the circuit input and the component degradation model is obtained.Next,the impact of component degradation parameters on circuit performance parameters is analyzed and the relationship is established.The circuit performance parameters containing component degradation information replace the parameters in the circuit behavior model,then the degradation behavior model of the circuit is obtained.The simulation of the model can be used to analyze the degradation behavior of the circuit.4.Design and implementation of circuit life prediction software.The software mainly implements three modules: component degradation analysis module,circuit component sensitivity analysis module,circuit degradation analysis and life prediction module.
Keywords/Search Tags:component degradation failure, sensitivity analysis, circuit degradation trend, circuit life prediction software
PDF Full Text Request
Related items