Simulation Of Steelmaking-Continuous Casting Process Based On Multi-Agent Technology

The two major directions for the upgrading of the steel industry towards manufacturing 2025 are green and intelligent,and it is still a very valuable issue to study the production process of the steelmaking and continuous casting production system.A more accurate intelligent simulation model can greatly improve the intelligence of the steelmaking production process.In view of the current steelmaking and continuous casting production process,this paper proposes a simulation modeling solution based on multi-agent technology,and conducts research on key technologies and multi-scenario analysis.The main contents and innovations are as follows:(1)In-depth study of the operation time of each equipment in the steelmaking production process,in which the LF furnace adopts the support vector machine multiple regression and Gaussian process regression methods according to the working conditions,and carries out a multivariate regression analysis.Based on the root mean square error of the obtained model(RMSE)is 0.043,which provides a more accurate prediction for the operating time of the LF furnace.At the same time,it is verified that the converter operation time is normally distributed,and the accurate prediction of the operation time can significantly improve the data accuracy of the simulation-based steelmaking production model.(2)Combined with the characteristics and difficulties of the steelmaking production process,a simulation modeling scheme based on multi-agent technology is proposed.From the perspective of behavior description,behavior trigger conditions,equipment selection principles,and interaction with other agents The simulation of the agent optimizes the model structure,and establishes a simulation model of the steelmaking production process based on the Anylogic multi-agent.(3)A universal multi-agent simulation platform is built based on Anylogic software.The model architecture can not only be used for the simulation of job plans,but also the simulation equipment can respond accordingly under various production disturbance conditions.By simulating and analyzing multiple scenarios such as furnace repair and equipment failure,the adaptation effect is verified.