CNN-BiGRU-based ultra-short-term power load prediction method

• Main Author: SUEYASU
• Format: Patent
• Language: Chinese; English
• Published: 06.09.2022
• 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 provides an ultra-short-term power load prediction method based on CNN-BiGRU. The ultra-short-term power load prediction method based on CNN-BiGRU comprises the following steps: step S01, preprocessing sample data; s02, converting the load influence factors and the historical load data into a matrix form; step S03, inputting the extracted data features into a BiGRU network to learn the data distribution and change rule of the BiGRU network; step S04, training a neural network; a step; and S05, prediction is carried out through the trained CNN-BiGRU network. According to the short-term load prediction method based on the CNN and the BiGRU, the influence of load influence factors in the historical period and the future period on the current prediction load is fully considered, meanwhile, feature extraction is carried out on the power load influence factors such as temperature, humidity and air pressure through the convolutional neural network, the feature extraction and the historical load are use