基于智能算法优化卡尔曼滤波器的电机转速估计

The Speed Estimation of Motor Using Intelligent Algorithm for Optimized Kalman Filter

针对感应电机扩展卡尔曼滤波器转速估计中难以取得卡尔曼滤波器系统噪声矩阵和测量噪声矩阵最优值的问题,提出了一种基于改进粒子群算法优化的扩展卡尔曼滤波器转速估计方法。算法通过融合遗传算法和粒子群算法的优点,采用可调整的算法模型对粒子群算法进行改进,将改进的粒子群算法对扩展卡尔曼滤波器中的系统噪声矩阵和测量噪声矩阵进行优化处理,将优化后的卡尔曼滤波器应用于感应电机转速估计,仿真实验表明,与试探法、标准粒子群算法及遗传算法比较,改进粒子群算法优化的扩展卡尔曼滤波器能够有效提高转速估计的精度,从而提高无速度传感器矢量控制系统的控制性能。

Aiming at the problem of getting the optimal value of system noise matrix and measurement noise matrix for extended Kalman filter（ EKF） which is widely used in induction motor speed estimation,a speed estimation method was presented by using improved particle swarm optimization（ IPSO）. By combining the advantages of genetic algorithm and particle swarm optimization, an adjustable algorithm was adopted in PSO. The EKF system noise matrix and measurement noise matrix was optimized by IPSO. Using the optimized EKF to estimate the speed of induction motor,simulation results show that the proposed method can effectively improve the speed estimation accuracy comparing with those obtained by trial and error methods,genetic algorithm（ GA） and standard PSO algorithm.