基于鲸鱼算法的电力系统能耗异常自动检测方法

An Automatic Detection Method for Abnormal Energy Consumption in Power Systems Based on Whale Algorithm

以往的电力系统能耗异常自动检测方法仅运用单一的检测方式,导致异常检测效果不佳,因此,本文提出了基于鲸鱼算法的电力系统能耗异常自动检测方法,通过分析大量历史数据,提取电力系统异常数据的特征。在此基础上,利用鲸鱼算法计算异常数据的属性权重,标注异常数据,运用物理故障异常数据检测和网络攻击异常数据检测这两种检测方式,实现能耗异常的自动检测。在实验测试中,与以往的电力系统能耗异常自动检测方法相比,本文提出的自动检测方法的异常检测准确率为94.18%,异常检测效果更好。

Previous automatic detection methods for energy consumption anomalies in power systems have had poor performance due to the use of a single detection method.Therefore,a whale algorithm based automatic detection method for energy consumption anomalies in power systems has been proposed.By analyzing a large amount of historical data,extract the characteristics of abnormal data in the power system.On this basis,the whale algorithm is used to calculate the attribute weights of abnormal data and annotate the abnormal data.Utilizing two detection methods,physical fault anomaly data detection and network attack anomaly data detection,to achieve automatic detection of energy consumption anomalies.In experimental testing,compared with previous automatic detection methods for energy consumption anomalies in the power system,the proposed automatic detection method has an anomaly detection accuracy of 94.18%,and the anomaly detection effect is better.