Optimization Of Operation Parameters Of Billet Reheating Furnace Based On Decision Model

Heating furnace is a typical industrial control system.Its heating process has strong parameter coupling,high nonlinearity and anti-interference ability.It’s the main energy in the iron and steel industry.With the continuous strictness of the heating process,the traditional control method cannot be very accurate,and to support this new theory.Aiming at the above problems in the production of 2250 mm rolling mill in a steel plant,an optimization design scheme based on multi-layer model matching is proposed to optimize process parameters,improve product quality,reduce energy consumption and improve the comprehensive competitiveness of products.Through a thorough analysis of the relevant theories,located with the complexity of the industrial process and the current optimization and control situation,a control idea based on modal matching is proposed,including: 1)establishing a comprehensive work index model;2)Fuzzy set theory and Euclidean distance transformation similarity method are used for multi-layer matching;3)RBF neural network is used to establish the prediction model,which provides the basis for the subsequent optimization algorithm 4)After obtaining the preconditions of the prediction model,the process parameters of the actual process are optimized by the optimization algorithm,and so on.to optimize the heating process operation parameters of the stepping heating furnace.The main work of this paper includes:(1)Based on the existing expert experience and historical data,a set of better operation model base is established,and a multi-layer model matching operation mode is proposed according to the actual situation.Through the determination of various parameters and data preprocessing,for the first model matching,the average clustering method shall be used,and then the euclidean distance shall be used to measure similarity,to determine the similarity between the current working state and the elements in the primary match,Complete the pattern matching process.This method meets the rapidity and accuracy of the actual production process to a great extent,increases the production efficiency to the greatest extent,reduces energy consumption and reduces the harm of industrial production to the environment.It is very important for the production process of walking beam heating furnace;(2)Before heating or under the specific process conditions of specific steel,it is difficult to find the best parameters due to insufficient data.Therefore,based on RBF neural network,RBF network is used to predict,and the model is improved to obtain the best operating parameters,which are updated to the best operating mode library to realize the expansion of the system.Matching the optimised process parameters to the above multi-level model and comparing the results of the simulation with the actual production situation show that the model matching development strategy is feasible and can provide a certain reference to the production of enterprises.