Research On Secondary Intelligent Control System Of Refining Furnace

With the fast growth of modern technology,industrial informatization and digitization have slowly become the tendency of corporate development,which can not only increase productivity,reduce production costs and improve product quality,but also strengthen business management and increase the satisfaction of customers.The secondary system of the refining furnace,as the key to the whole process of steelmaking,enables automated monitoring and control of the production process in order to improve quality management and coordinate the rhythm of steelmaking.This system takes a 120T refining furnace under a steel company as the research background and designs a secondary intelligent process control system for the refining process,which mainly realizes the functions of model guidance and optimization calculation,which lays the foundation for the realization of digital twin refining furnace.Firstly,this paper focuses on the basic process and characteristics of LF refineries,and establishes a mathematical model for the metallurgical process of refineries.Among them,the temperature prediction model is mainly due to the fact that the refining furnace can not find a suitable temperature measuring element to achieve the purpose of continuous temperature measurement at this stage,and the current molten steel temperature is usually obtained by multiple temperature measurements to adjust the arc heating power.The alloy feeding model mainly requires the alloying operation of molten steel in the refining process,which is usually calculated by artificial empirical algorithm,which fails to consider the selection of appropriate material ratio in the case of the lowest alloy cost.Secondly,in response to the above issues,this paper establishes a temperature prediction model based on S VR algorithm by analyzing the energy balance and transfer process of refining furnace.in order to make the model operator converge as rapidly as possible,PSO algorithm is introduced to optimize the above model.Through comparative analysis,the error of PSO-SVR regression prediction model is relatively small,and the furnace times with prediction errors within ±5℃ account for 94%of the total furnaces.The model can realize the continuous prediction of molten steel temperature.In order to meet the end point and outbound temperature,guide the basic level to select arc heating,and reasonably control the heating power to provide data support and reference.Aiming at the operation of molten steel alloying,the alloy yield is optimized by the average furnace method.On the premise of meeting the steel composition standard,an alloy batching model based on LP algorithm is established with the lowest alloy cost as the goal,and finally the optimal solution set is obtained.The alloy composition adjustment is completed automatically by guiding the basic level,so as to achieve the purpose of reducing the alloy input cost.Then,the system architecture and communication design were carried out for the entire refining secondary process control system.The overall framework of the system is based on the design pattern of three-tier architecture,divides and designs the modules according to the business requirements and functions,and finally realizes the processing and recording of process data in the refining process level system,such as feeding,temperature measurement,argon blowing,electrification,wire feeding and so on.In the aspect of communication,as a processlevel system,users with different identities and responsibilities are designed to interact with the basic level and the management level respectively.Among them,OPC communication technology is used to exchange data with the basic level to achieve real-time data acquisition and guidance.On the other hand,the management level uses MQ queue communication technology to send and upload data,with the help of the complete working mode and reliable working characteristics of the message queue to ensure the stability and security of the communication system,and finally realize the reasonable scheduling of the steelmaking process.Finally,according to the functions of each module of the refining secondary system,a friendly man-machine interface is designed,including main control function information,steel test control information,LF smelting details,LF smelting process information,temperature model prediction and alloy model guidance.The two-level intelligent control system of refining furnace designed in this paper is based on process theory and aims to improve the technical level of refining furnace equipment,combined with related database technology and communication technology to design and implement the system.The system guides the actual production through the established process model,which is of great practical significance to improve the enterprise economy and ensure the refining efficiency.