Study On Temperature Control Of Regenerative Reheating Furnace Based On Neuron Network

Regenerative heating furnace is a new type of energy-saving heating furnace,which adopts the method of high temperature and low oxygen content.It is a kind of environmentally friendly heating furnace.However,because of the testing means of the instrument and diversity of heat transfer in reheating furnace,the instrument cannot directly and accurately detect the temperature of billet and the temperature distribution in the furnace.A compound method based on BP neural network control PID is proposed in this paper.By offline learning and classifying and identifying the model of the control object,it can adjust itself gradually to meet the characteristics of the control system object by correcting the weights of the network structure.Because the furnace temperature object of heating furnace is a controlled object,thus the temperature of regenerative heating furnace can only be controlled by neural network PID.When adjusting the furnace temperature,the perfect control index cannot be obtained by using the PID data obtained from the above model.The furnace temperature control is a nonlinear and multivariable control system.At present,most of the combustion control methods are PID combined with double cross limiting.but its shortcomings are also outstanding,it can only achieve the optimal combustion state in a stable state.In light of this characteristic of the temperature of heating furnace,the method of training PID with neural network is used to control the temperature of heating furnace.The given temperature of heating furnace can be adjusted at any time according to the actual situation of the system,so as to meet the technological temperature requirement of billet.In order to realize the function of fitting any non-linear function peculiar to the automatic control neural network of heating furnace,to make it control the optimal combination of P,I and D parameters by self-learning of the system.Figure22;Table1;Reference 52...