Research on Statistics and Anomaly Detection of Monitoring Information for Power Transmission and Transformation Equipment Under Big Data Environment
A monitoring signal parameter statistical analysis method based on isolated forest algorithm is proposed for the detection of abnormal status data of power transmission and transformation equipment,and a centralized monitoring auxiliary decision-making system is designed.Firstly,an analysis was conducted on existing data anomaly detection methods,and the shortcomings in existing research on status data detection of power transmission and transformation equipment were pointed out.Secondly,the overall architecture of monitoring information data collection and statistics was provided,and the data statistics precautions for key links such as the control technology support system(D5000),dispatch management system(OMS),and online monitoring system of the smart grid were analyzed.Then,the implementation principle of the isolated forest algorithm was presented,and three anomaly data detection models were constructed using LOF and K-means algorithms.Through simulation,it was found that the anomaly data detection model based on the isolated forest algorithm has higher accuracy and computational speed.Finally,a case study was conducted on a 500 kV transmission line.The results showed that the proposed centralized control assisted decision-making system can accurately identify equipment abnormal states.
big datamonitoring signalspower transmission and transformation equipmentparameter statisticsisolated forest algorithm
NETL
NSTL
万方数据
