Research on Multi-Time Scale Self-Evolving Control Strategy of Industrial Load Considering Rebound Effect

In the context of global energy transition, the penetration of renewable energy in the power system is increasing, and grid scheduling is facing unprecedented challenges. As an important adjustable resource, industrial loads have great potential in peak shaving and valley filling, and balancing power supply and demand. In this paper, we focus on the multi-timescale optimization problem of industrial load participation in grid scheduling, and propose an orderly peak shaving, multi-timescale self-evolutionary optimal scheduling strategy that considers the participation of multiple industrial loads. The strategy analyzes the regulation characteristics of cement, electrolytic aluminum, and steel industry loads, and builds a multi-timescale scheduling model by combining their peak shaving potentials and rebound effects. In the day-ahead scheduling, intra-day rolling scheduling and real-time scheduling, this paper establishes the corresponding optimization models, and designs the objective functions and constraints for the load characteristics of different industries. In order to cope with the dynamic changes of the grid operation environment, this paper also introduces an incremental learning algorithm, which uses real-time data to update the model with self-evolution to improve the adaptability and intelligence of the scheduling strategy. Finally, the effectiveness of the proposed strategy is verified by case analysis, which shows that the strategy can effectively reduce the peak shaving cost while improving the flexibility and reliability of grid scheduling.