Research On Flatness Recognition And Control Of Cold Rolled Strip Steel Based On Cloud Network

The rapid,stable and sound development of the iron and steel industry has led to the rapid growth of plate and strip products.At the same time due to the gradual improvement of the economy,the improvement of the living conditions of the residents,so the demand for plate and strip is also increasing.In today's society,people on the quality of strip products in improving,but in terms of plate rolling and controlled by rolling process,detection and control system,how to improve the quality of the plate has become a hot and difficult research in the world in the field of rolling.On the basis of T-S cloud inference network,this paper designs a T-S cloud inference flatness control system based on particle swarm optimization algorithm.Firstly,according to the popular artificial intelligence technology,with the development of it,a new method to control the shape of the cold strip mill is proposed.The method solves the problem of low precision in the traditional method.The model is based on T-S neural network.It has the characteristics of strong anti-interference effect of the traditional neural network,at the same time,it also takes into account the advantages of cloud model to deal with the uncertainty of data.Secondly,the T-S neural network is applied to the flatness recognition,and the pattern recognition model based on particle swarm optimization is established.The simulation results show that the T-S neural network is higher than the T-S fuzzy neural network.This has proved that the method compared with the traditional cloud neural network is a new method for flatness recognition of a better recognition effect,through PSO and GA for the optimization comparison of particle swarm optimization T-S neural network recognition cloud reasoning results identify GA optimization is closer to the actual shape based on the network.Finally,based on the T-S cloud inference neural network,the flatness prediction model is established.Based on the T-S cloud inference network,the 4 input and 2 output T-S.The design of the flatness control system on this basis,and the flatness control system for a 900 HC six roller reversible cold rolling mill,By comparing the simulation results can be seen through the shape on the shape optimization of PSO controller provides higher recognition accuracy and control accuracy is also improved,therefore the effectiveness of the proposed control system is verified.