往复式压缩机的虚拟可靠性试验方法研究

A Virtual Reliability Test Method for Mechanical Multi-body Stochastic System

针对往复式压缩机传动机构的疲劳失效问题,提出一种适用于机械多体随机系统的虚拟可靠性试验方法。考虑到环境载荷、尺寸误差及材料强度等因素的随机性,根据乘同余法编写随机数子程序,利用Pro/E软件和APDL语言的参数化功能实现虚拟样机的随机抽样;基于显式动力学算法,采用ANSYS/LS-DYNA程序进行动态数值仿真分析,获得虚拟试验数据;根据K-S法分析数据的分布类型,考核关键零部件各处结构的可靠性,从而找出整个系统的薄弱环节。同时,依据传统的疲劳强度理论,建立可靠性分析模型,分析传动机构的可靠性。通过两种分析方法的比较,说明了新方法的合理性,为机械多体随机系统的故障预报和最优维护提供前提准备。

Considering fatigue failure of the reciprocating compressor＇ s drive mechanism, a virtual reliability test method for mechanical multi-body stochastic system was proposed. Regarding randomness of the environ- ment load, size error and material strength and according to subroutine written with multiplicative congruential method, both Pro/E software and APDL language＇ s parameter function were used to realize random sampling of the virtual prototype; basing on explicit dynamics algorithm, the ANSYS/LS-DYNA program was used in the dynamic numerical simulation to obtain virtual test data; through analyzing the data distribution with K-S method, the reliability of each key part was examined to find out loophole of the whole system ; at the same time, according to the traditional fatigue strength theory, the reliability analysis model was established to ana- lyze reliability of the drive mechanism. Comparing two analytical methods can prove new method＇ s rationality. This provides the premise for mechanical multi-body system＇ s fault prediction and optimal maintenance.