Power load prediction method and prediction system

Abstract

The invention provides a power load prediction method and prediction system. The prediction method comprises the steps that S1, dividing a power load curve into N classes based on power load data; S2,selecting three groups of data sets, optimizing a penalty parameter C and a kernel function parameter g by utilizing a grid search method based on cross validation, and training N (N-1) / 2 SVM classifiers to obtain a classification model for classifying power load curves; S3, classifying the power load data by using a classification model; and S4, predicting the power load based on the classifiedpower load data and the load prediction model. The power load curves are classified, and the classification model based on the SVM classifier is established, so that load prediction is realized on thebasis of classification of the power load curves, the parameters C and g in the classification model are optimized, and the prediction accuracy of the classification model is greatly improved.

Description

technical field [0001] The invention relates to the field of electric power technology, in particular to an electric load forecasting method and a forecasting system. Background technique [0002] Power load forecasting is a series of forecasting work on power load. From the perspective of forecasting objects, electric load forecasting includes forecasting of future demand for electricity (power), forecasting of future electricity consumption (energy) and forecasting of load curves. Its main work is to predict the time distribution and spatial distribution of future power load, and provide reliable decision-making basis for power system planning and operation. [0003] In power load forecasting, many factors affect the predicted value of power load to varying degrees. Some factors vary due to nature, such as weather, and some vary according to regional conditions, such as the speed of industrial and agricultural development; some factors are not estimated. Events, such as ...

AI Technical Summary

Problems solved by technology

[0004] However, most of the existing power load forecasting methods are based on empirical comparison of past data, and the forecasting results are not accurate enough.

Images