基于改进矩阵束的高压直流次同步振荡检测

Detection of Subsynchronous Oscillation in HVDC Transmission System Based on Improved Matrix Pencil Method

矩阵束是一种可以用来进行振荡模态参数辨识的算法。鉴于传统的矩阵束算法在有噪声情况下模态阶数辨识效果不理想,提出将信息熵引入矩阵束进行改进,进而将改进的矩阵束算法用于高压直流的次同步振荡模态的辨识。然后与改进Prony法和矩阵束算法对比分析,证明了基于信息熵的矩阵束算法的有效性。仿真结果表明,此算法可以快速、准确地辨识出次同步振荡模态参数,且该算法抗噪声能力强,可用于电力系统次同步振荡在线检测。

Matrix pencil algorithm is such an algorithm by which the parameter identification of oscillation modal can be conducted. In view of dissatisfied result of modal order identification by traditional matrix pencil algorithm in noisy environment, it is proposed to lead information entropy into matrix pencil to improve the latter and apply the improved matrix pencil to the identification of subsynchronous oscillation （SSO） occurred in HVDC power transmission system. The effectiveness of the information entropy based matrix pencil algorithm is proved by comparative analysis on the identification results respectively obtained from the improved matrix pencil algorithm and the improved Prony algorithm under noisy environment. Simulation results show that the proposed algorithm can identify modal parameters of SSO rapidly and accurately, and in view of its strong ability of anti-noise it could be available to use the proposed algorithm in on-line detection of SSO.