Analysis Of Static And Dynamic Reliability Of Mechanical Parts Based On Stochastic Finite Element Method

In the traditional mechanical components reliability analysis, usually directly according to the concrete expression of the stress to get the probability model, at the same time combination with strength probability model establish interference model to reliability analysis. But in the practical, mechanical parts often cannot give specific expression, led to the obstacles of reliability used widely in practical engineering. In the aspect of reliability data statistics, through direct do reliability data statistics is the high cost of test, and due to the different structure of parts lead to stress distribution difference greatly, the results are often not universal. As a result, mechanical components reliability research focus has been around the reliability model and the reliability data two aspects.In this paper, on the basis of stochastic finite element theory, using ANSYS probability analysis module to reliability research. In terms of the static reliability analysis, by Monte Carlo stochastic finite element, the response surface stochastic finite element and the mixed stochastic finite element reliability do the static reliability analysis for mechanical parts, solve the problem of no show expression parts reliability modeling analysis, contrast three kinds of stochastic finite element method get their own characteristics and conditions of use at the same time. In the aspect of dynamic reliability analysis, with the stochastic finite element static reliability analysis of stress data as the foundation, using the Monte Carlo method for dynamic reliability simulation analysis, solve the problem of lack of dynamic reliability analysis data. Specific research contents and results are as follows:1)Through the establishment of industrial robot APDL forearm reliability analysis file,choose the Monte Carlo stochastic finite element method do the structure static reliability analysis on the forearm, get the static reliability of the forearm and the parameter sensitivity analysis results. Show that Monte Carlo stochastic finite element method can well solve parts static reliability analysis problem that show no expression, at the same time is also a kind of good commonality stochastic finite element reliability analysis method.2)In robot forearm, for example, further using the response surface stochastic finite element method for analysis. Through to the fitting of the response surface of a large number of Monte Carlo simulation quickly, get results of static reliability analysis of forearm. Show that the response surface to fitting structure instead of forearm finite element model of response surface stochastic finite element method can be analyse quickly, has a higher efficiency analysis than the Monte Carlo method. Through detailed contrast two kinds of stochastic finite element method, finally get the respective applicable conditions of static reliability analysis of Monte Carlo stochastic finite element and response surface stochastic finite element.3)Comprehensive Monte Carlo characteristics and response surface two stochastic finite element method, in combination with parameter sensitivity analysis get mixed stochastic finite element reliability analysis method. In parameter and structure are relatively complex engine connecting rod as an example, detailed comparison of the mixed stochastic finite element, the response surface stochastic finite element and the Monte Carlo stochastic finite element three static reliability analysis methods, results show that the mixed stochastic finite element method in the parts of the complex structure and input parameters in the static reliability analysis, has higher computational efficiency than other two methods.4)When parts dynamic reliability analysis, to support the plate as an example, using the static reliability analysis results of Monte Carlo stochastic finite element, extract a large number of samples of the stress data that suffix for.pdrs. Further obtained the stress distribution characteristics of dynamic reliability analysis by MATLAB distribution fitting, to establish a dynamic reliability interference model. Final edit Monte Carlo dynamic reliability simulation program to analyze the dynamic reliability of the support plate, get change with load frequency curve of dynamic reliability. The results show that based on the stochastic finite element method to solve the dynamic reliability data lack problem is achievable, provides a new reference method for dynamic reliability analysis of mechanical parts.