Dynamic reliability of mechanism based on direct probability integral method

Reliability of mechanism systems is very important. However, actual engineering machinery systems are affected by various random factors, such as limited manufacturing accuracy, unavoidable assembly errors, essential motion clearance, etc., which cause the actual motion of the mechanism system to deviate significantly from its ideal design, resulting in a significant deficiency in the dynamic reliability of the mechanism system. Therefore, in this paper, we start from the general dynamic equation of the mechanism system to study the highperformance computational method of dynamic response reliability variation due to parameter uncertainty. Firstly, the multi-body dynamics equations of the mechanism system with parameter uncertainties are established. Secondly, the variable weight sampling point strategy is used to substitute the sampling point into the multi-body dynamics equation to obtain the corresponding dynamic time-domain response. Then, the direct probability integral method (DPIM) is used to calculate the extreme value reliability of the dynamic time-domain response. Finally, a typical computer numerical control (CNC) machine tool is taken as an example to verify the effectiveness of the dynamic reliability method proposed in this paper for general mechanism systems. The main advantage of this method is that it eliminates the need for repeated derivations of dynamic models for different mechanism systems and can handle reliability problems with complex performance functions, thus offering greater generality. In addition, it solves the problem of the time-variant reliability of mechanical systems. In particular, the use of low-discrepancy sequences as the initial point set further improves the computational efficiency of the dynamic reliability analysis of the mechanism system.