Natural gas demand combined prediction method and system based on weight adjustment strategy

• Main Authors: ZENG ZHIMEI, LYU XING, YU ZEHUI, HUANG JUNJIE, ZHANG BOTING, ZHANG XUEJING, ZHANG DABIN
• Format: Patent
• Language: Chinese; English
• Published: 07.03.2023
• 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 natural gas demand combined prediction method and system based on a weight adjustment strategy. The method comprises the following steps: acquiring natural gas demand data; dividing the data into a training set, a verification set and a test set; respectively training the ten single models through the training set; determining an initial combination weight of the model through a traditional combination prediction weighting method; an offset error is defined to measure the difference between the prediction effect of each single model point and the overall prediction effect, a weight adjustment optimization method is proposed based on the offset error, and the weight of each time point single model of the verification set is adjusted; and taking an adjustment weight mean value under 40% of the time span as a combination weight of each time point of the test set to realize natural gas demand quantity combination prediction. According to the method, more reasonable weights can be determin