基于MESCM算法的电网异常状态检测改进方法

An Improved Method for Abnormal State Detection of Power Grid Based on MESCM Algorithm

为了提高现有电网异常状态检测方法的计算效率,扩大其应用范围,基于样本协方差矩阵最大特征值(maximum eigenvalue of sample covariance matrix,MESCM)算法,提出一种改进算法。该算法基于随机矩阵理论,以标准M-P律的上确值作为样本协方差矩阵特征值的阈值,并引进动态的异常状态检测阈值,实现对不同信噪比环境下电网异常状态的检测分析。案例分析以IEEE 39节点网络和一个实际的高压直流输电系统为模型,通过设置短路故障,将改进MESCM算法与平均谱半径(mean spectral radius,MSR)算法和传统MESCM算法进行对比分析。结果表明,相较于MSR和MESCM算法,改进MESCM算法具有计算速度快、抗干扰能力强和普适性高的优势。

In order to improve the calculation efficiency and expand the application scope of existing methods for abnormal state detection of power grid,an improved algorithm based on maximum eigenvalue of sample covariance matrix(MESCM)algorithm is proposed in this paper.The improved MESCM algorithm is based on the random matrix theory.The upper value of standard M-P law is used as the threshold of the eigenvalues of sample covariance matrix,and a dynamic threshold of abnormal state detection is introduced to detect and analyze the abnormal state of power grid under different signal-to-noise ratio environments.Based on IEEE 39-node network and an actual high voltage direct current(HVDC)transmission system,the improved MESCM algorithm is compared with the mean spectral radius(MSR)algorithm and the traditional MESCM algorithm by setting short-circuit faults.The comparison results show that the improved MESCM algorithm has the advantages of fast calculation speed,strong anti-interference ability and high universality.