Research on Power Market Supply and Demand Balance and Optimization of Regulating Power Supply and Energy Storage Price Mechanism Based on Deep Learning

With the high proportion of new energy access, the pressure of system regulation continues to rise. The unpredictability of electricity market price forecasting, the complexity of dispatching and coordination, and the insufficient economic performance of energy storage have become key issues that need to be addressed. This paper uses the power load and electricity price data disclosed in 2024 to build a linkage modeling architecture that integrates deep learning prediction, dispatching optimization and energy storage pricing mechanism. Construct an electricity price prediction model with LSTM and Transformer as the core components to achieve high-precision prediction on an hourly scale; formulate a nonlinear supply and demand optimization model based on the forecast results, taking into account the minimization of power generation costs and load fluctuation control; adopt the energy storage marginal cost model and price elasticity function to introduce a dynamic bidding mechanism for energy storage and regulating power sources. The final results of the experiment show that this approach can significantly reduce the cost of system operation, suppress price fluctuations, and improve the response sensitivity and revenue magnitude of energy storage resources, showing good practical performance and engineering compatibility.