Energy data anomaly reason analysis method based on random forest and support vector machine

• Main Authors: LI PENGCHENG, ZHU CHUANJING, HU XIAONAN, XU JUN, ZHANG HAITAO, GUO ZHENGXIONG, HU HAOHAN, ZHANG LI, ZHANG MINGSHANG
• Format: Patent
• Language: Chinese; English
• Published: 24.12.2021
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses an energy data anomaly reason analysis method based on a random forest and a support vector machine. The method comprises the following steps: 1, carrying out data cleaning processing, data tagging and model parameter tuning by adopting a model training module; 2, supporting to read related power load control abnormal data from MySQL, Oracle and Postgre by adopting an abnormal data receiving module; 3, carrying out data processing, wherein key information screening, redundant data deletion and time window calculation are conducted on abnormal data through the big data technology, and wherein the Python is used for extracting the power load control abnormal data from the Oracle and the Postgre in a JDBC mode; and 4, performing reason analysis on the abnormal data by adopting an abnormal reason analysis module, and feeding back an abnormal reason analysis result. The method is based on a random forest and a support vector machine model, and performs reason analysis on abnormal data belon