Model-data hybrid driven bilateral electricity market electricity price prediction method

Abstract

The invention discloses a model-data hybrid driven bilateral electricity market electricity price prediction method. The method specifically comprises the following steps: firstly, providing a bilateral market data set sampling method based on nuclear network sampling; then establishing a data regression model by using an optimization method by means of a basic idea of data driving, and reversely solving value estimation of a single market subject through mutual constraint among multiple market subjects; the price of the negotiation contract which does not occur is predicted by using the determined value estimation parameter; and finally, screening different prediction models generated according to different data sets to obtain a final prediction model. According to the invention, the price of the contract which is not reached in the electricity market can be predicted, and reference can be provided for operation supervision of the market and making of own power generation and power utilization plans by a market subject.

Description

technical field [0001] The invention belongs to the field of electricity price forecasting in the electricity market, and in particular relates to electricity price prediction in a bilateral electricity market contract under the condition of insufficient market data types and data volumes, specifically a model-data hybrid-driven electricity price prediction method in a bilateral electricity market. [0002] technical background [0003] The electricity market is one of the important means to achieve the goal of "carbon neutrality and carbon peak". Since the "new electricity reform" in 2015, China has continued to promote the reform of the power market. In 2020, the direct transaction power of the national power market will reach 2.476 billion kWh, accounting for 32.9% of the total power consumption of the whole society. The power market has gradually become an electric energy supply- Main approaches and technical solutions to be matched. Bilateral trading is one of the most ...

AI Technical Summary

Problems solved by technology

In addition, bilateral contracts rely on the contracting relationship between the buyer and the seller. Due to the homogeneity of electricity commodities, this relationship is usually difficult to maintain. Market players often change contracting partners in the next delivery period, and the contracting relationship shows a strong randomness. sex and chance

This results in a short duration of buyer-seller contracts, a low renewal rate, and it is difficult to form a long time series

The above-mentioned characteristics of the two-sided market make it difficult to apply the traditional electricity price forecasting method, and the direct analysis of the price itself also loses a stable carrier. Therefore, it is necessary to find other indirect carriers and a relatively stable expression relationship between the indirect carrier and the electricity price.

Images