Short-term power consumption load prediction method

Abstract

The invention discloses a short-term power consumption load prediction method, which comprises the steps of collecting historical environment data and historical load data of short-term power consumption, preprocessing the collected data, and dividing the data into a training set sql1 and a test set sql2 of the historical short-term power consumption data; building a GRULight GBM (Geographic BlockManagement) model, wherein the input quantity of the GRU neural network model is short-term power consumption environment data, the output quantity of the GRU neural network model is GRU predicted power consumption load Ep1, the input quantity of the LightGBM model is the short-term power utilization environment data and the GRU predicted power utilization load Ep1, and the output quantity of theLightGBM model is the LightGBM predicted power utilization load Ep2; adopting a training set sql1 to train the GRU-LightGBM model; adopting a test set sql2 to verify that the model training is completed; inputting the power utilization environment data of the to-be-predicted power utilization load period into the trained GRU-LightGBM model, and taking the outputted LightGBM prediction power utilization load Ep2 as a prediction result of the short-term power utilization load. According to the method, a LightGBM and GRU model fusion method is adopted, various environmental influence factors areintegrated, and the short-term power consumption load can be accurately predicted.

Description

technical field [0001] The invention relates to the field of electric load forecasting, in particular to a short-term electric load forecasting method. Background technique [0002] Power load forecasting refers to determining the power load value at a specific time in the future based on the known power demand, considering political, climate, social and other related factors, and meeting certain accuracy requirements. Improving load forecasting technology is conducive to planned power consumption management, formulating a reasonable power grid operation mode and unit maintenance plan, thereby improving the overall economic and social benefits of the power system. Therefore, power load forecasting has become an important part of realizing the modernization of power system management. [0003] Power load forecasting can be divided into medium and long-term load forecasting and short-term load forecasting according to time. Medium- and long-term load forecasting refers to loa...

AI Technical Summary

Problems solved by technology

At present, there are many methods for short-term residential electricity load forecasting, including gray theory method, time series method, regression analysis method, support vector machine method, neural network method, etc. To meet production needs, new technologies are needed to break through the limitations and achieve better forecasting results

Images