考虑误差非独立性的电力系统参数辨识估计

Power System Parameter Identification and Estimation Considering Measurement Error Dependencies

确定量测量间的误差传递特性是建立精确量测模型的重要前提。实际量测系统中量测量由所有单相遥测数据构成,在各母线上分析量测误差间的相关性,计算符合遥测数据实际采集特征的量测误差协方差阵。提出了考虑量测误差相关性的电力系统参数辨识估计,利用加权残差率绝对值求和的方法辨识出参数误差支路,采用改进增广状态量法进行参数估计,逐一修正参数误差,利用状态估计结果验证参数辨识估计准确性。IEEE算例仿真结果表明,所提出的算法较传统参数估计结果更接近于系统真值,同时提高了状态估计的精度。

The definition of measurement error transfer characteristics among measurement values is an important prerequisite for a precise measurement model. On the basis of three-phase measurement configuration in an actual system,the dependencies among each measurement at the same node are analyzed accurately in this paper. The measurement error covariance matrix suitable for the actual measurement system is calculated. And power system parameter identification and estimation considering measurement error dependencies are proposed and branches with parameter errors are identified by taking advantage of weighted maximum residual ratios. And parameter estimation is carried out by using augmented state algorithm to improve the accuracy of the estimators. The test results of IEEE standard systems show that the parameter estimation results produced with the proposed method are closer to the true values,which also improves the accuracy of the state estimation.