Research on the Risk Quantification Model of the Connection between Power Trading and Spot Market in the Medium and Long Term

This paper centers on the power load peak-shaving control model under the risk of the day-ahead spot market, with the aim of optimizing the operational efficiency of the power system and the smooth evolution of the market mechanism. Through an in-depth analysis of the current state of the medium and long-term power market in Liaoning, this paper uncovers the linkage issues present in crucial aspects such as the composition of market participants, trading categories, pricing mechanisms, particularly the coordination challenges between the medium and long-term trading and the spot market. This paper innovatively formulates an algorithmic model grounded in machine learning, capable of forecasting the trend of load variations while considering multiple facets such as the power supply architecture, congestion administration, settlement tactics, and more, to achieve flexible peak load balancing of the power system and precise alignment of market settlements. Subsequently, this paper employs a deep neural network to tackle intricate load forecasting tasks, integrating historical data alongside real-time market information to enhance forecasting accuracy. Moreover, by incorporating a risk assessment module, the model can effectively pinpoint and quantify the uncertainties of the spot market, offering decision support for the power load participation system. To validate the efficacy and applicability of the model, this paper conducts a meticulous simulation analysis based on the operational data of the Liaoning power grid. The findings indicate that the proposed model not only considerably boosts the peak load capacity of the power system but also efficiently diminishes the risk associated with market settlements, providing robust underpinning for the wholesome operation and enduring advancement of the power market.