摘要
全阶磁链观测器是以状态重构的方式来获取感应电机不可测量的相关信息,因其数学模型为四阶微分方程组,给数字化算法实现研究和分析带来难度,并且采用不同的离散化方法对系统控制性能的影响不同.基于数值积分单步法原理,分别采用一阶精度的欧拉法和二阶精度的梯形法,并采用适用于特殊应用场合的高精度三阶龙格-库塔法和四阶龙格-库塔法,实现以定、转子磁链为状态变量的开环和闭环全阶磁链观测器的数字化算法.针对2.2 kW感应电机实验平台,通过大量的实验结果验证不同算法的有效性、精度及其执行效率.
Some important information of the induction motor which cannot be measured directly can be obtained by state reconstruction. However, the mathematical model of a full order observer is a four order differential equation, which is not easy to analysis and digitally implemented. Different discretization methods have different effects on the system performance. Four methods of full order observer digital realization are proposed based on the single footwork principle in numerical integration. These are first- order Euler method, second-order Trapezoidal method, and third-order and fourth-order Runge-Kutta methods. Validity, efficiency and precision of the four methods are shown with results of experiments on a 2.2 kW induction motor platform.
出处
《上海大学学报（自然科学版）》
CAS
CSCD
北大核心
2012年第6期582-588,共7页
Journal of Shanghai University:Natural Science Edition
基金
国家自然科学基金资助项目(51107079)
关键词
感应电机
全阶磁链观测器
速度自适应
离散化
矢量控制
采样步长
induction motor, full-order flux observer, speed adaptive, discretization, vector control,sampling step
