基于自适应卡尔曼滤波的负荷参数在线辨识方法

Online Identification Method of Load Parameters Based on Adaptive Kalman Filter Algorithm

传统负荷模型参数辨识方法在数据获取、辨识精度和在线辨识上受到限制,且基于传统卡尔曼滤波的参数辨识算法易受噪声干扰,辨识结果不稳定。为此提出一种基于自适应卡尔曼滤波的负荷参数在线辨识算法。首先建立线性化的负荷模型,基于广域测量系统(WAMS)同步相量测量单元(PMU)测得的实时在线数据,运用预报误差法的思路,使用改进的Sage-Husa自适应卡尔曼滤波算法辨识负荷模型参数。基于浙江电网220 kV华金变电站PMU数据的算例表明了自适应卡尔曼滤波算法辨识结果具有更高的参数稳定性和良好的拟合度。

The traditional parameter identification method of load model is limited in data acquisition,identification accuracy and online identification.Moreover,the parameter identification algorithm based on traditional Kalman filter is easy to be disturbed by noise and the identification results is unstable.Therefore,an on-line identification algorithm of load parameters based on adaptive Kalman filter was proposed.Firstly,the linearized load model was established.Based on the real-time on-line data measured by the synchronous phasor measurement unit(PMU)of wide area measurement system(WAMS),the improved Sage-Husa adaptive Kalman filter algorithm was used to identify the load parameters with the prediction error method.The case study based on PMU data of 220 kV Huajin substation in Zhejiang power grid shows that the identification result of adaptive Kalman filter algorithm has higher parameter stability and good fitting degree.