Power Prediction of Iron and Steel Enterprises Based on Improved Conditional Hidden Semi-Markov Model

As a typical energy-intensive industry, the iron and steel industry has a large demand for electricity. Power forecasting for iron and steel enterprises is conducive to power grid demand management and power quality assurance. The equipment functions of iron and steel enterprises are different and need to be coordinated. The influence of physical parameters on power prediction in the production process cannot be ignored. Therefore, based on the improved conditional hidden semi-Markov model (CHSMM), this paper proposes a power prediction model for steel industry users considering the physical parameters of the production process. Firstly, the influence of physical parameters such as material quality and process temperature drop on power prediction is analyzed, and then the power prediction model is established based on CHSMM. On this basis, considering the coordination between the production processes of the enterprise, an improved CHSMM considering the influence of the above physical parameters is constructed. The results show that the proposed method can improve the prediction accuracy of user power in the iron and steel industry.