基于相对值法和Hurst指数的电网停电事故自相关性分析  

Self-correlation Analysis of Power Grid Blackouts Based on Relative Value Method and Hurst Exponent

作　　者：于群 屈玉清 石良 

机构地区：[1]山东科技大学电气与自动化工程学院,山东省青岛市266590 [2]国网山东省电力公司莱芜供电公司,山东省莱芜市271100

出　　处：《电力系统自动化》2018年第1期55-60,124,共7页Automation of Electric Power Systems

基　　金：山东省自然科学基金资助项目(ZR2016EEM13)~~

摘　　要：为从宏观和全局角度揭示中国电网停电事故的内在动力学机理,分析了中国电网停电事故的相关性。首先,基于相对值法并结合R/S方法对中国电网1981—2014年停电事故损失负荷序列进行了相关性分析,指出了中国电网的停电事故损失负荷序列具有强的长程相关性,对其相关性方向进行了判定,并依据Hurst指数大小对其相关性程度进行了有效量度。然后,随着时间的推移,分析了中国电网停电事故损失负荷相关性的变化情况,揭示了中国电网停电事故损失负荷正相关性会越来越强、相关性程度会越来越高的规律。最后,在宏观层面上,预测出中国电网未来几年的大停电事故的损失负荷绝对值有继续增大的趋势,而损失负荷相对值有继续减小的趋势。To reveal the internal dynamics mechanism of blackouts in power grid of China from both the macro and global perspective, the correlation of blackouts in power grid of China is analyzed. Firstly, based on the relative value and the R/S method, the power loss correlation of blackouts in power grid of China from 1981 to 2014 is researched. It is proved that the power losses of blackouts in power grid of China have the strong long-range correlation, to determine the direction of correlation, and to effectively measure the degree of correlation according to the Hurst exponent. Secondly, over time, the change of the power loss correlation of blackouts in power grid of China is analyzed. It is revealed that the positive correlation is becoming stronger while the degree of correlation is becoming higher than the power losses of blackouts in power grid of China. Finally, on the macro level, the trend of the absolute value continues to increase and relative value continues to decrease for the power losses of large blackouts in power grid of China in the next few years can be predicted.

关 键 词：停电事故 相对值法 自相关性 R/S方法 HURST指数 趋势预测 

分 类 号：TM711[电气工程—电力系统及自动化]