一种高精度APNCKF算法在谐波检测中的应用

为了进一步提高含噪环境下谐波检测的精确度,提高卡尔曼滤波器的稳定性,对系统噪声协方差进行了分析,通过不断的在线辨识出过程噪声协方差,提出了一种自适应过程噪声协方差卡尔曼滤波算法。该算法利用序贯最大化可信度更新先验信息来辨识过程噪声,然后通过卡尔曼滤波器进行迭代运算,估计出相应的幅值和相位。该算法最大的特点就是辨识出的过程噪声Q的骤然增大匹配的即是谐波幅值暂降的出现。通过在MATLAB环境下进行谐波仿真验证,结果表明该算法在准稳态条件下较好地跟踪电力系统谐波状态,且与常规卡尔曼、基于最大似然准则的卡尔曼、小波/小波包变换相比,该自适应算法的收敛速度较快、滤波精度高、实时性以及稳定性较好,具有重要的工程实际意义。

Application of High Precision APNCKF Algorithm of Power Grid Harmonic Signal Detection

In order to improve the accuracy of harmonic detection in the noisy environment and improve the stability of the Kalman filter, the covariance of the system noise is analyzed. By continuously recognizing the process noise covariance continuously, an adaptive process Noise covariance Kalman filter algorithm is presented. Firstly, the algorithm uses prioritized maximization of confidence to update prior information to identify process noise, Then iteratively calculates the amplitude and phase through Kalman filter. The biggest feature of this algorithm is that the sudden increase of the identified process noise Q is the occurrence of the sag of the harmonic amplitude. The simulation experimental results show that the proposed algorithm can track harmonic state of power system under quasi-steady state condition better than conventional Kalman, Kalman based on maximum likelihood criterion, wavelet / wavelet packet Compared with the transform, the adaptive algorithm has faster convergence speed, higher filtering precision, better real-time and stability, and has important practical significance.