Short-term daily load prediction method considering steel futures prices and real-time electricity price factors

Abstract

The invention provides a short-term daily load prediction method considering steel future price and real-time electricity price factors. The method comprises the following steps of determining a correlation coefficient among the steel future price, the real-time electricity price, the date type and the load by using a covariance analysis method, and determining the steel future price, the real-time electricity price and the date type value in the short-term load prediction input data by using the correlation coefficient based on the load value; performing standardization processing on the input data, and establishing a short-term load prediction input data training set; based on the short-term load prediction input data training set, training the long-term and short-term memory model of load prediction to obtain a short-term load prediction model with appropriate parameters; and inputting the load values in the test set into a short-term load prediction model with appropriate parameters to complete short-term daily load prediction of the load with impact. Steel futures are used as input quantities, and errors of short-term daily load prediction caused by impact loads are reduced.

Description

technical field [0001] The disclosure relates to the technical field of electricity load forecasting of power grid dispatching department, in particular to a short-term daily load forecasting method and system considering steel futures price and real-time electricity price factors. Background technique [0002] Power system load forecasting is based on the historical data of power load, economy, society, weather, etc., to explore the influence of the change law of the historical data of power load on the future load, and to seek the internal relationship between the power load and various related factors, so as to predict the future. Scientific prediction of power load. In power system planning and design, operation management and power market transactions, it is necessary to have an accurate forecast of load demand changes and load characteristics. This is the important reason why people continue to study and develop the theory of power system load forecasting. [0003] W...

AI Technical Summary

Problems solved by technology

[0010] 1. The time series method and the multiple linear prediction method are mostly aimed at the linear relationship model, and the prediction error is large

[0011] 2. The above method has the disadvantages of slow convergence speed and low prediction accuracy in dealing with large data volume load forecasting

[0012] 3. The above method does not consider the special influencing factors of the impact load, and the accuracy is not high

Images