A method, system, computer equipment and storage medium for electricity price probability prediction

Abstract

The present invention provides an electricity price probability prediction method, system, computer equipment and storage medium. The method includes: according to the electricity price data in the training period, using a prediction model to predict and obtain the first electricity price prediction data; according to the first electricity price prediction data, Using distribution model fitting to obtain the error characteristics of the first electricity price forecast data; using a clustering algorithm to cluster and analyze the error characteristics of the first electricity price forecast data to obtain the central characteristics of the first electricity price forecast data; according to the target time period electricity price data, Obtaining second electricity price prediction data by using a prediction model; correcting the second electricity price prediction data according to the central characteristics of the first electricity price prediction data to obtain electricity price probability prediction data in a target time period. The embodiment of the present invention not only ensures the scientific accuracy of electricity price prediction and the analysis accuracy of error probability, but also takes into account the feasibility of using the prediction results for risk analysis, which is of great significance to the pricing decision-making and risk assessment of power suppliers.

Description

technical field [0001] The invention relates to the field of power grid electricity price prediction, in particular to a method, system, computer equipment and storage medium for electricity price probability prediction based on error clustering. Background technique [0002] At present, the first batch of eight power spot market construction pilot projects determined by the National Development and Reform Commission and the National Energy Administration have all entered the trial operation stage, and my country's power market construction has achieved another important breakthrough. With the gradual deepening of market-oriented competition in the power industry, the settlement of the electricity wholesale market based on the nodal electricity price theory will change with market conditions, and the volatility of electricity prices will become increasingly obvious. How to make scientific and accurate electricity price forecasts and give It is very important for market playe...

AI Technical Summary

Problems solved by technology

[0003] The existing domestic and foreign electricity price prediction research results mainly focus on the prediction of price certainty, and rarely involve its uncertainty or risk research. The main reasons are as follows: On the one hand, it is difficult to model periodic social activities, and it is easy to introduce electricity price Prediction error, and the overall consideration of error does not reflect the characteristics of electricity price errors in different periods; on the other hand, the predicted electricity price data will be applied to large-scale optimization problems such as risk assessment, and considering too many risk scenarios will introduce a lot of risk Analysis workload

In practical applications, if the electricity price forecasting method provided can only guarantee the relative accuracy of electricity price forecasting, and only conduct relatively simple error analysis, but cannot take into account the analysis accuracy of error probability and the feasibility of forecasting results for risk analysis, the other Difficult to be directly used in pricing decisions and risk assessments for generators

Images