Power consumption prediction method, device, computer equipment and storage medium

Abstract

The present application relates to a power consumption prediction method, device, computer equipment and storage medium. The method includes: obtaining the power consumption data sequence; performing multi-scale wavelet decomposition on the power consumption data sequence to obtain decomposed data with different frequency characteristics; extracting characteristic data corresponding to the decomposed data, splicing the characteristic data to obtain the power consumption data The eigenvector of the sequence; the eigenvector and the corresponding additional prediction feature are vector spliced, and the power consumption prediction is performed based on the spliced ​​vector, and the power consumption prediction result is obtained. Through the above-mentioned power consumption forecasting method, the instantaneous fluctuation change of the data is captured while modeling the periodic signal, thereby improving the accuracy of the power forecasting result.

Description

technical field [0001] The present application relates to the field of electric power technology, in particular to a power consumption prediction method, device, computer equipment and storage medium. Background technique [0002] The stable supply of electric energy is the key to ensuring the development of social economy and the normal life of the people. At the same time, accurate electricity consumption forecasting can provide reliable guidance for the production and dispatch of electric energy. For the electricity consumption data collected by the terminal, it generally appears as periodic data with certain fluctuations. Data from different sources are periodic and have different statistical characteristics. For example, the periodicity of electricity consumption by urban residents is not obvious, On the other hand, electricity consumption in commerce, manufacturing, culture, education and entertainment has less volatility and relatively obvious periodicity. The power ...

AI Technical Summary

Problems solved by technology

The power consumption forecast needs to consider the different statistical characteristics of the power consumption data from different sources, and it also reflects its volatility while capturing the periodicity, which brings challenges to the power forecasting task.

[0003] At present, the commonly used models in the task of electricity consumption forecasting are difficult to capture the instantaneous volatility of the data while modeling the periodicity of the electricity consumption data. The commonly used methods try to model the periodicity and fluctuation of the electricity consumption data in an implicit form. However, this method has the problem of easy over-fitting data and it is difficult to capture the instantaneous fluctuation of electricity consumption data, resulting in low accuracy of electricity consumption prediction

Images