SCADA data cleaning method for wind generating set and electronic equipment thereof

• Main Authors: LIU MINGDA, LI CHENG, DENG YI, LIU YAN, WANG ZHENHUA, CAO MENGNAN, LIN LIN, YANG XIAOCHU, YANG RONGCHANG, MA ZHIYUAN
• Format: Patent
• Language: Chinese; English
• Published: 05.09.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention provides a wind generating set SCADA (supervisory control and data acquisition) data cleaning method and electronic equipment thereof, and the data cleaning method is based on unsupervised learning and comprises the following steps: S1, data points in a non-power-generation wind speed interval are cleaned; s2, clearing shutdown data points; s3, eliminating limited power data points; s4, data points with wrong monitoring signals are cleared; and S5, data points of suspected limited power and fault states are cleared according to a clustering algorithm. According to the method, an algorithm based on K-Means unsupervised learning is adopted, and data points of suspected limited power and fault states can be identified. Therefore, the data points with poor quality can be effectively eliminated, and an effective data set is provided for unit state evaluation calculation. A power curve formed by the data cleaned by the data cleaning method is closer to a theoretical power curve, and data points with p