高速列车传动系统特征参数经典域优化

Optimization of Classical Domains for High-Speed Train Transmission System

为监测高速列车传动系统的运行状态,根据可拓学理论,建立了传动系统各部件的运行状态物元,提出了一种部件正常运行状态下的特征参数经典域优化方法.利用部件样本集与其正常运行状态之间的最大综合关联度构建了适应度函数,并利用并行粒子群优化算法进行解算,确定了特征参数的经典域范围.与用数理统计方法得到的经典域结果进行了对比分析,结果表明,用本文经典域优化结果得到的最大综合关联度的最大值和平均值分别提高了3.63%和2.51%,经典域优化结果更符合部件的实际运行状况.

In order to monitor the operating condition of a high-speed train transmission system, matter elements that present the operating condition of components of the transmission system were established by the extension theory, and an optimization method for determining the classical domains of characteristic parameters of the matter element that presents a component＇s normal operating condition was proposed. In this method, a fitness function was constructed using the maximum comprehensive correlative degree （MCCD） between a component＇s sample sets and its normal operating condition. Then, the parallel particle swarm optimization （PSO） algorithm was adopted to solve the fitness function and determine the classical domains. In addition, the optimization results of classical domains were compared with those determined by the statistical method. The results show that the maximum and average of MCCDs obtained from the optimization results of classical domains are improved by 3.63% and 2.51% , respectively, which demonstrates that the optimization results of classical domains more conform to components＇ actual operating conditions.