Energy System Prediction and Optimal Scheduling Based on Operating Condition Identification for Steel Enterprises

In the steel production process, there is significant energy generation and consumption. Achieving flexible and efficient utilization of energy is crucial for steel enterprises to achieve energy savings and cost reduction. The integrated energy system of a steel enterprise involves various energy sources and production equipment. Understanding the relationship between equipment energy production and consumption, as well as the coupling relationships between different energy sources, is key to realizing flexible and efficient energy utilization. This paper establishes an integrated energy system prediction and optimization scheduling model based on operating condition identification, while considering time-of-use electricity pricing, with the goal of maximizing economic performance. The model also accounts for production processes, energy coupling, safety requirements, and time-of-use electricity pricing. To evaluate the model's accuracy and applicability, a case study was conducted. The results show that by using an improved K-means clustering algorithm for operating condition tracking and identification, the gas holder level prediction accuracy exceeded 95%. Additionally, the frequency of steam turbine generator load fluctuations was reduced, lowering adjustment costs for the units. The fluctuation amplitude of gas holder levels was also minimized, keeping the levels within the optimal operating range. After the model's implementation, the self-generation rate of the steel enterprise increased, the proportion of purchased peak electricity decreased, and significant economic benefits were achieved.