基于凸模型的复杂系统不确定分析

Complex System Uncertainty Analysis Based on Convex Models

研究了输入响应关系较为复杂系统的不确定性分析。采用凸模型来描述不确定参数,将系统分解为若干个输入响应较为简单的子系统,结合泰勒级数,推导了系统响应的近似计算公式,适用于小范围不确定性问题;对于大范围不确定性问题,采用超长方体单元剖分,将其分解为若干小范围不确定问题,确定了系统响应的范围,然后再采用近似分析,提高了问题求解精度。对某六连杆功能发电机的不确定分析结果表明该方法有效可行。

The convex model theory is applied in the uncertain analysis of complex systems. A complex system is decomposed into several subsystems firstly, then the intervals of system responses is estimated using Taylor series when the uncertainty level is low. While the larger uncertainty problem is referred to, it is decomposed into several smaller uncertainty problems using hyper- cube. The method is verified by example of a six- link function- generator linkage . A good approximation for the unceaain intervals is shown by result.