基于多目标进化算法的异步电动机现场效率测算

介绍了一种基于多目标进化算法（MOEAs）的异步电动机现场实时效率测定方法。通过对多目标算法进行优化、比较,提出使用非支配排序遗传算法Ⅱ（NSGA-II）和强度帕累托进化算法2（SPEA2）的低侵入式方法用于异步电动机效率估算,仅需电动机运行时通过传感器检测其实时转子速度和定子电阻,而无需拆下电动机或单独做一些实验项目来获取所需参数。通过5.5kW电动机的实践表明,该方法在估算异步电动机效率方面是有效的,尤其在常规的负载范围内,用该方法的估算值与实际试验值的误差小于3%;相互比较后发现,NSGA-Ⅱ方法的估计结果略优于SPEA2方法的结果。

In-situ Efficiency Measurement of Induction Motor Based on MOEAs

A method based on multi-objective evolutionary algorithms (MOEAs ) for real-time and in-situ efficiency determination of induction motor is introduced. The multi-objective algorithm is optimized, and it introduces a low-intrusive level method based on non-dominated sorting genetic algorithm-II( NSGA-II ) and strengths Pareto evolutionary algorithm2( SPEA2)  for efficiency estimation of the induction motor. Therefore, the equivalent circuit method, and the segregated losses method combine with MOEAs. Through the practice of a 5.5 kW motor, it shows that, the method on the estimation of induction motor efficiency is effective, and in the normal range of the load, the estimating value and the actual value of test error is less than 3%. And compared to each other, it is found that NSGA-Ⅱresults are slightly superior to that of the SPEA2.