Online PMU data error correction method and system based on LSTM model

• Main Authors: ZHU XIN, ZHENG LING, LIU MINGZHONG, TENG YUFEI, DING XUANWEN, WU JIE, ZHANG XU, DAI YUHAN, ZHOU WENYUE, LONG CHENG, SUN YONGCHAO, ZHANG CHUN, CHEN FEI, LI XIAOPENG
• Format: Patent
• Language: Chinese; English
• Published: 06.03.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses an online PMU data error correction method and system based on an LSTM model. The method comprises the following steps: carrying out normalization processing on PMU historicaldata under normal operation of a power grid; training an LSTM neural network by using the normalized PMU data and establishing a prediction model; normalizing the current PMU data, inputting the normalized PMU data into a prediction model, and predicting the PMU value of the power grid in the next time period; comparing the error of the prediction result relative to the measured value on line, determining error data according to a set error threshold, and replacing the error data with prediction data. Real-time correction of PMU data is realized. 本发明公开了本发明提供一种基于LSTM模型的在线PMU数据纠错方法及系统，所述方法包括：对电网正常运行下的PMU历史数据做归一化处理；利用归一化后的PMU数据训练LSTM神经网络并建立预测模型；将当前PMU数据归一化后输入预测模型，预测电网下一时段的PMU数值；在线比较预测结果相对于实测值的误差，根据设定的误差阈值确定错误数据，对于错误数据采用预测数据更换；实现对于PMU数据实时修正。