Lifetime Prediction Of MEMS Butterfly Resonator Gyroscope Based On Multi-stress Accelerated Test

Electromechanical systems are usually used in multiple stresses,and have multiple failure modes.However,the traditional accelerated test method based on single stress and single failure modes is mainly for electronic products and mechanical components,so it is difficult to accurately and efficiently evaluate the reliability of electromechanical systems.Therefore,it is urgent to study accelerated test method with multiple stresses and reliability modeling method with multiple stresses and multiple failure modes to evaluate the reliability level of electromechanical system.MEMS gyroscope usually serves in temperature,vibration,rotation,electrical stress and other comprehensive conditions,and has high reliability and long life requirements.Therefore,there is an urgent need to study the multiple stresses acceleration test method of MEMS gyroscope and the reliability evaluation method of MEMS gyroscope,which can provide a basis for the optimal design of MEMS gyroscope reliability.In order to solve the requirements of MEMS reliability evaluation,this thesis aims to establish the degradation model of MEMS butterfly resonantor gyroscope(BRG)by failure mechanism analysis,to study the acceleration test method and reliability modeling method with multiple stresses,and to provide theoretical method and technical support for MEMS reliability evaluation and lifetime prediction.The main reserch contents and innovation points are as follows:1.The degradation model of MEMS BRG was studied.The influence of residual stress in the silicon microstructure on the resonant frequency of the excitation modal was studied by an analytical expression.The finite element method is used to carry out the prestress modal analysis of the silicon microstructure with the condition of different residual stresses(29 groups),and the finite element simulation results is close to the simulation results of the proposed analytical model.The degradation model of resonance frequency and degradation model of mechanical sensitivity are established based on the analytical model.A long-term static test of MEMS BRG was designed and carried out.The test results show that the resonant frequency and mechanical sensitivity of the MEMS BRG gradually decrease with time,which is consistent with the numerical simulation results of the degradation model.2.The degradation model of bias instability and an acceleration factor model under temperature stress of MEMS BRG were established and verified by experiments.Firstly,the noise term of MEMS BRG is mainly composed of angle random walk,bias instability and rate random walk based on Allan variance analysis.Secondly,we found bias instability increases linearly with time based on the analysis of the degradation test data,and the degradation rate of the bias instability is bigger with the increase of the temperature stress level.Finally,the remaining life of the MEMS BRG was predicted.3.The existing acceleration model consists of few stress terms,lacks stress coupling terms,and multiple parameters are difficult to be estimated.A new universal multiple stresses acceleration model and multiple stresses acceleration factor model were established based on the classic Arrhenius model.Then a novel multiple parameters estimation method of multi-stress acceleration model was proposed based on particle swarm optimization and maximum likelihood estimation method.Finally,six simulation cases and a practical application case were used to verify the correctness and accuracy of the proposed method.4.Considering the coupling effect of multiple stresses on life characteristics,and the dependence of the degradation mode and the random shock mode,a new reliability model with N stresses M failure modes was established based on the universal multiple stresses acceleration model and dependent competing model.A statistical inference method of accelerated degradation testing with multiple stresses and dependent competing failure modes was proposed.A two-step analysis method was proposed to estimate unknown parameters.Finally,a practical example was used to demonstrate accuracy of the proposed model and method.5.In order to provide a complete engineering application case for the multiple stresses accelerated test method,we used it to predict the lifetime of MEMS BRG,which is a key component of a micro-inertial navigation system.Firstly,a multiple stresses accelerated test platform was constructed,which can simultaneously supply multiple stresses such as vibration,temperature,and voltage for MEMS BRG.Then,the multiple stresses accelerated test plan was designed based on the theory of uniform and orthogonal experiments,and the multiple stresses accelerated degradation test of MEMS BRG was implemented.Finally,the reliability model of MEMS BRG with multiple stresses was established.In summary,with the funding of the "13th Five-Year Plan" pre-research project,this thesis establishes degradation models of key performance indicators for the MEMS BRG,establishes an accelerated test method with multiple stresses,and proposes a reliability modeling and statistical analysis method with multiple stresses and multiple failure modes.The research results of this thesis can provide theoretical foundation and technical support for the reliability evaluation and lifetime prediction of other high-reliability,long-lifetime electromechanical systems.