Temperature Model Research Of Regenerative Reheating Furnace Based On Hybrid Modeling Technique

Regenerative reheat furnace is the latest energy saving and environment friendly reheat furnace as the development of HTAC in the 1980s, comparing with heat exchange reheat furnace, it takes the advantages of higher temperature in air(gas) pre-heat up, while lower temperature in smoke venting, as well as higher thermal efficiency etc. At present, the regenerative reheating furnace is obtained widespread application in the world scope. The reheating'furnace is not only one of the important equipment, but also a large energy consumer on steel rolling line. Only reasonable control of billet's temperature and its distribution can assure rolling quality and reducing energy consume in reheating furnace.but refer to billet temperature field, the inner temperature of billet now can't be reached or measured by using some instruments easily. After billet come out of furnace, the workers just know whether billet temperature is eligible. As a result, billet temperature can't be traced and it can not be retrieved if the temperature is unfit. In order to predict billet temperature ahead, a general measure is to establish model of billet temperature, by which optimal control based on billet temperature model is realized to improve the billet heating quality.In this paper, based on walking beam regenerative reheating furnace system, there are a brief overview of regenerative reheating furnace and algorithm foundation of Solution modeling question. According to the analysis of the mechanism used in the pastî–¯nd intelligent modelling methods of modelling and hybrid modelling methods, this paper presents one hybrid model of reheating furnace based on neural network. physical mechanism model is established with total heat exchange method in the hybrid model firstly. In order to get inputs of neural network, this paper uses PCA to make measured model based on the data which can be acquired easily from a walking beam regenerative reheating furnace system. By gathering some steel mill the field data, the simulation is carried on to the model. In view of simulation result, the method of this paper and other heating furnace modeling methods has been compared. Meanwhile accuracy of modeling is validated through the data, and further the on-line compensated and proofread strategy of the model is also proposed, so the model to be more precise. The best furnace temperature setting of heating furnace is proposed using GA algorithm, make sure the goal of the quality requirement and the energy saving is achieved.The simulation result indicated that the billet temperature can be quite accurately calculated by the billet temperature model. The model in this paper is showed its excellence on comparing with other modeling method. This paper uses the neural network model for approaching the difference between the actual system and normal model to improve the model's approaching and generalization ability. This result is a beneficial attempt for study of steel mathematical model for the future.