Font Size: a A A

Thickness Control Of Hot-rolled Steel Strip Based On BP Neural Network

Posted on:2016-04-02Degree:MasterType:Thesis
Country:ChinaCandidate:P WangFull Text:PDF
GTID:2271330470479905Subject:Materials Science and Engineering
Abstract/Summary:PDF Full Text Request
In this paper,the basic principle and method of thick control on hot rolled strip are reviewed.The thick controlling system of Chao Yang Steel Company 1700 mm ASP of An Steel Group Corporation are introduced. In order to improve the hot rolled product thickness control precision, a research and analyze of the existing production technology and production data was made,based on Chao Yang Steel 1700 ASP in An Steel group co., LTD., Using BP neural network to predict the strip thickness deviation and the neural network PID controller is designed,and made comparison of deviation forecasting and AGC control curve with traditional thick control system.At the same time, it is discussed that the BP neural network design number of hidden layer, the layer number of neurons learning algorithms.It is constructed that hot rolled strip steel thickness AGC model and PID controller.,based on the BP neural network.According to the result of off-line simulation analysis shows that the hot rolled strip steel thickness control system based on BP neural network is superior than traditional thickness control system. In terms of model prediction deviation of network training, it can be obtained that the grouping optimization of same hierarchy with BP neural network is more accurate than the whole optimization of same hierarchy with BP neural network.In terms of AGC PID controller designing, it can be obtained that PID controller based on neural network is more accurate than the traditional hydraulic control precision of PID controller, it can realize the PID controller on the P/I/D value of effective automatic adjustment.It can be seen according to the implementing the results that the accuracy control(plus or minus 50 microns) can achieve 99.5% or more of Hot rolled strip steel thickness control system which based on BP neural network.Because of the actual application of the BP neural network in hot-rolling strip thick control is very little, Thus based on BP neural network of hot rolled strip steel thickness control online application can effectively improve the level of strip steel thickness control, its prospect is broad, can increase significant economic benefits for the enterprise.
Keywords/Search Tags:Hot-rolling strip, BP neural network, PID Controller
PDF Full Text Request
Related items