论文部分内容阅读
为改善柔性机构动态可靠性分析的效率和精度,基于支持向量机(Support Vector Machine,SVM)回归理论,提出了一种柔性机构动态可靠性分析高效率高精度的SVM回归极值法(SVM Regression Extremum Method,SREM)。首先,介绍了柔性机构可靠性分析的基本理论;其次,结合蒙特卡罗法(Monte Carlo,MC)和SVM回归理论,建立了柔性机构动态响应极值的代理模型,并利用代理模型进行了柔性机构的可靠性分析;最后,以柔性夹紧机构的可靠性分析为例,利用SREM加以验证。结果表明:SREM的计算时间约为蒙特卡罗法的20%,远远少于蒙特卡罗法;SREM的计算精度几乎与蒙特卡罗法保持一致,当可靠度大于98%时,SVM回归极值法的计算精度与蒙特卡罗的计算精度完全一致。
In order to improve the efficiency and accuracy of dynamic reliability analysis of flexible mechanism, a new method based on Support Vector Machine (SVM) regression theory is proposed to analyze the dynamic reliability of flexible mechanism. The high efficiency and accuracy SVM regression method (SVM Regression Extremum Method, SREM). First of all, the basic theory of reliability analysis of flexible mechanism is introduced. Secondly, the agent model of flexible mechanism dynamic response extremum is established by combining Monte Carlo (MC) and SVM regression theory. Reliability analysis of the mechanism. Finally, the reliability analysis of the flexible clamping mechanism is taken as an example and verified by SREM. The results show that the calculation time of SREM is about 20% of that of Monte Carlo method, far less than that of Monte Carlo method. The calculation accuracy of SREM is almost the same as that of Monte Carlo method. When the reliability is more than 98% The calculation accuracy of the value method is exactly the same as that of Monte Carlo.