以测点正常率最大为目标的状态估计改进算法

发布时间：2018-05-25 01:35

本文选题：状态估计 + 测点正常率　；参考：《电力系统自动化》2014年11期

【摘要】：已有研究指出基于测量不确定度的测点正常率是真值未知情况下可信的状态估计结果评价指标,进而提出了以测点正常率最大为目标的抗差状态估计新方法。该方法通过整数测点评价函数的连续近似,建立了状态估计的非线性连续变量优化模型,并采用内点法求解该模型,获取近似优化的估计结果。在上述研究基础上,文中提出了一种改进算法:通过对异常测点进行相容性校验,以获得测点正常率更高的估计结果;通过考虑不等式约束的最小二乘估计对正常测点进行滤波,以获得精度更高的估计结果。算例表明该改进算法可有效提高估计结果的抗差性和准确性,且增加的计算量能够满足应用要求。
[Abstract]:It has been pointed out that the measurement point normal rate based on the uncertainty of measurement is a reliable evaluation index of the state estimation results in the case of unknown true value, and a new robust state estimation method aiming at the maximum normal rate of the measured point is proposed. The nonlinear continuous variable optimization model of state estimation is established by the continuous approximation of the evaluation function of integer measurement points. The interior point method is used to solve the model and the estimated results of approximate optimization are obtained. On the basis of the above research, an improved algorithm is proposed in this paper: by checking the compatibility of the abnormal points, we can get the higher estimation results of the normal rate of the measured points; In order to obtain more accurate estimation results, the normal measurement points are filtered by the least square estimation with inequality constraints. Numerical examples show that the improved algorithm can effectively improve the robustness and accuracy of the estimation results, and the increased computational complexity can meet the requirements of the application.
【作者单位】：国家电力调度控制中心;清华大学电机工程与应用电子技术系;电力系统及发电设备控制和仿真国家重点实验室清华大学;
【分类号】：TM744

【参考文献】