Residential electricity consumption prediction method based on historical similar days and probability statistics and application thereof

• Main Authors: LI PENG, YANG MENG, TIAN CHUNZHENG, ZHANG YIHAN, LI HUIXUAN, ZHENG YONGLE, ZU WENJING, YANG QINCHEN, ZHANG HONGKAI
• Format: Patent
• Language: Chinese; English
• Published: 25.07.2023
• Subjects: CALCULATING; 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 a residential electricity consumption prediction method based on historical similar days and probability statistics and application thereof. The method comprises the following steps: constructing a similarity feature vector; calculating a similarity factor between the prediction day and the historical day; preprocessing historical data; modeling a power consumption behavior prediction model; modeling a load electrical characteristic model; and modeling a load demand prediction model. According to the method, probability statistics is carried out by using historical similar daily power consumption data, the residential power consumption behavior is analyzed and predicted, a power consumption demand prediction model from bottom to top is constructed, the problem of data selection is solved through data preprocessing, the prediction accuracy is improved, and the defects in current research are made up. 一种基于历史相似日和概率统计的居民用电预测方法及其应用，包括构建相似度特征向量；计算预测日与历史日之间的相似度因子；历史数据预处理；用电行为预测模型建模；负荷电气特征模型建模