Research On LF Smelting Data Acquisition And Alloy Feeding Control System

This paper researches the resource waste and inefficiencies problems,existing in manual control alloy components in LF(Ladle Furnace)secondary smelting,to obtain a set of alloy feeding optimization control system.Taking into account the alloy feeding control system performance depends on established accuracy of the elements yield predictive model and alloy feeding model.So first of all,according to actual situation and the needs of users of the original steel smelting integrated automation system,establishes using OPC technology and Socket technology for smelting process data acquisition and communications technology route,and uses C#to complete software design in Win7+Oracle environment,which lays a reliable data base for alloy feeding control system;then establishes elements yield prediction model for multiple linear regression,multiple nonlinear regression and BP neural network based on field data obtained and compares them respectively,synthetically considering model accuracy and operational stability requirements,determines using multiple nonlinear regression method to model element yield prediction model;finally for steel production of mills,by analyzing the traditional alloy feeding model and researching alloy feeding model based on simplex methoduses dual simplex method to solve alloy feeding model based on simplex method,then obtains a satisfactory alloy addition and total cost of alloy,achieves alloy feeding control system ultimately.Data acquisition system in this paper has been applied to the production site,simulation results based on field data demonstrate the effectiveness of the proposed method.