Monthly power consumption prediction method based on capacity utilization hours and Gaussian distribution

Abstract

The invention discloses a monthly power consumption prediction method based on capacity utilization hours and Gaussian distribution. The method comprises the steps: firstly calculating the transformercapacity utilization hours of each power user in each historical month, and calculating the mean value of the transformer capacity utilization hours of each historical month; then, a histogram statistical method is adopted to obtain the frequency of occurrence of the mean value of the transformer capacity utilization hours in different statistical intervals; secondly, obtaining typical sample points through Gaussian distribution, and constructing an industry typical capacity utilization hour curve according to the typical sample points; and finally, predicting monthly power consumption of theindustry by using the hour curve and the total capacity of the transformer of the user to be predicted according to the typical capacity. According to the method, the electricity capacity utilizationhours are calculated, a typical curve of the capacity utilization hours is obtained through strict histogram statistics and Gaussian distribution calculation, the data which can most reflect the electricity utilization level is mined essentially, and the prediction accuracy is improved.

Description

technical field [0001] The invention relates to a monthly power consumption prediction method based on capacity utilization hours and Gaussian distribution, and belongs to the technical field of monthly power consumption prediction. Background technique [0002] my country's economy has entered a new normal, and the transfer of industrial structure puts forward new requirements for power forecasting, which greatly increases the difficulty of power forecasting. Conventional analysis and prediction based on natural growth rate can no longer meet the requirements of the development of the times, and it is urgent to explore a new prediction mechanism to more scientifically predict monthly electricity consumption. [0003] The classic monthly electricity forecasting methods include the unit consumption method of output value, the electricity consumption elastic coefficient method, the load density method, the growth rate method, the per capita electricity method, etc.; the tradit...

AI Technical Summary

Problems solved by technology

However, most of these methods are only based on the electricity consumption itself, or comprehensively consider the prediction ideas of other external factors, and the accuracy is poor.

Images