Electricity consumption prediction method

Abstract

The invention discloses an electricity consumption prediction method, which comprises the following steps that: obtaining electricity consumption data and relevant impact factors; calculating the electricity consumption data and the relevant impact factors according to a copula function to obtain rank correlation coefficients; sorting the rank correlation coefficients corresponding to the impact factors according to a descending order, and selecting a preset quantity of impact factors which rank in the front as effective impact factors; or selecting the impact factor corresponding to a threshold value the which exceed a preset relevant coefficient as an effective impact factor; and inputting the effective impact factor and the electricity consumption data into a pre-constructed predictionmodel to be trained, and obtaining an electricity consumption prediction result. By use of the electricity consumption prediction method, through the copula function calculation, the rank correlationcoefficients are obtained to realize the selection of the impact factors, and a calculation amount can be lowered while prediction accuracy is improved, prediction speed is improved, and through the prediction model, a prediction process can be simple and efficient. Therefore, the method can be used for more accurately and efficiently predicting the electricity consumption, and the prediction accuracy of the electricity consumption is improved.

Description

technical field [0001] The present invention relates to the technical field related to electricity consumption prediction, in particular to a method for electricity consumption prediction. Background technique [0002] As an important basic industry of the national economy, the electric power industry has an important impact on the development of the national economy and the improvement of the quality of life of the public. The research on power consumption structure analysis and development trend modeling is particularly important for the development of the power industry. It is related to the rational optimization of the power grid layout of related enterprises, the scientific dispatch of power between regions, and the formulation of power demand. In the process of discussing the structure of electricity consumption, the analysis of annual electricity consumption is a very important part. At present, the contradiction between supply and demand of electricity has become inc...

AI Technical Summary

Problems solved by technology

However, the factor variables considered in this model are too few to show the impact of external variables on social electricity consumption. In addition, because the impact of external factors on electricity consumption is nonlinear, the gray relational degree has problems in describing the nonlinear correlation. insufficient

[0009] From the introduction of the above literature, we can know that the current prediction of electricity consumption involves few external factors and correlation coefficients of electricity consumption, and few external factors are considered. Most of them use linear correlation coefficients, so for the actual situation recovery is poor

Images