Integrated electricity consumption increase prediction method and device combined with high-frequency influence factors

Abstract

The invention discloses an integrated power consumption increase prediction method and device combined with high-frequency influence factors, and the method comprises the steps: S1, selecting factors which may affect the power consumption increase as alternative influence factors, constructing an XGBoost model based on the alternative influence factors, measuring and calculating the importance of each alternative influence factor on the power consumption increase, and calculating the importance of each alternative influence factor on the power consumption increase; determining influence factors used for constructing the prediction model; S2, constructing a plurality of XGBoost models as a single prediction model based on the influence factors obtained in the step S1; and S3, predicting the power consumption by using the plurality of single prediction models constructed in the step S2 to obtain a plurality of groups of prediction values, calculating a mean square error value by using the real power consumption and the prediction values, and integrating the prediction values of the plurality of single prediction models by using the mean square error value to obtain a final prediction result.

Description

technical field [0001] The invention relates to the technical field of power system load analysis, in particular to an integrated electricity consumption growth prediction method and device that integrates high-frequency influencing factors. Background technique [0002] Electricity is the basic energy for social development and the core element of the entire national energy strategy. Accurate prediction of electricity consumption growth is an important guarantee for reliable and efficient operation of power systems. In recent years, with the profound changes in the economic and social situation at home and abroad, the optimization and adjustment of the industrial structure in various parts of my country, and the acceleration of the transformation of new and old kinetic energy, the future growth of electricity consumption will be affected by more external factors, and the uncertainty of the growth trend will increase. In order to adapt to the rapidly changing economic and s...

AI Technical Summary

Problems solved by technology

Previous studies only considered the use of a single model to construct a linear or nonlinear time series model, and the forecast of electricity consumption growth was based on only one modeling result. The single model’s insufficient fitting or generalization ability will lead to low accuracy of the forecasting model

Images