基于不确定测度的电力系统抗差状态估计(一)理论基础

Uncertain Measure Based Robust State Estimation of Power System Part One Theoretical Principle

传统电力系统状态估计的理论基础是传统统计学的大数定律,即当量测量的数目趋近于无穷大时,估计值以概率为1逼近于真值。而实际系统中量测量的数目有限,有时甚至是小样本,此时传统状态估计的评价指标和估计精度没有理论上的保证。针对这种情况,引入不确定理论体系下不确定测度概念,阐述了在状态估计中不确定测度与测量不确定度和量测误差的区别与联系;进一步在不确定理论体系下,重新给出了正常测点、异常测点、测点正常率的定义,并提出考虑正常率和量测量偏离真值程度的新的状态估计结果评价指标,证明了其合理性。

The theoretical basis of traditional power system state estimation is the law of large numbers based on traditional statistics, that is, when the number of measurements approaches imrinity, the estimated value approximates to the true value with 1 probability. However, the number of measurements in the actual system is limited and sometimes even small, there is no theoretic guarantee for the evaluation index and estimation precision of the traditional state estimation. In view of this situation, the paper introduces the concept of uncertain measure under the uncertainty theory system, and explains the difference and relation among the uncertain measure, uncertainty of measurement and measurement error in the state estimation. Furthermore, the normal measurement, abnormal measurement and normal rate are redefined under the uncertainty theory system. And a new evaluation index of state estimation result considering the normal rate and the measurement deviation degree from the true value is proposed and the rationality is proved.