Research On The Modeling Of Boiler Turbine Coordination System Based On Deep Neural Network

Coordination control system is the core unit of automation control system in thermal power plant.Its control performance will directly affect the long-term safe,stable and economic operation of the unit.Accurate model of coordination system is the precondition of control system algorithm design and performance research,so it is very necessary to establish an accurate model of the coordinating system of machinery and furnace.In this paper,on the basis of in-depth study of neural network algorithm,to deal with the shortcoming of too many artificial parameters in the neural network,the distribution of data can not be accurately reflected and easily fall into local optimum trap,this paper proposes to increase the layers of the neural network,improve the learning algorithm,and improve it into the deep confidence network of the deep neural network.The actual non-linear objects are modeled by using neural network and deep belief network respectively.According to the simulation results of MATLAB,the model of deep belief network is more accurate.Finally,based on the analysis of the dynamic and static characteristics of the coordinating system,the deep belief network is applied to model and test the coordinating system.The final simulation results show that the deep belief network has strong learning ability and generalization ability,and the three-in-three-out parametric model of the coordinated system achieves high fitting accuracy,which can correctly reflect the dynamic characteristics between the input and output of the coordinated system,and achieves the desired results.It provides an important basis for subsequent control algorithm design based on coordination system model.