Modeling And Simulation Of Boiler Combustion Process Based On The Neural Network And PSO

The energy structure of a country has a great impact on the consuming of itsresources. China has very rich coal resources which leads to that the powerstations in China are mainly based on coal power. Boilers provide thermal andelectric power to both industrial production and our living. The boiler system is acomplex thermal power engineering system which has multiple-input andmultiple-output which is essential to the industry.Combustion is a very complex process which could be affected by manyvariables and it has characteristics of nonlinear, long delay, strong coupling. Anddue to the use of time, machine wear degree, coal and other reasons, the originalset parameters of the boiler have been unable to meet the control requirements,therefore using a large amount of data to build a real model of the power stationbased on the neural network is a good method, this paper takes a single-inputsingle-output system as the object. Subtractive Clustering Method isselected to ascertain the number of neurons in hidden layer in order toavoid the shortcomings that the traditional method has to do a lot ofexperiment which could affects the project efficiency and PSO is adoptedto optimize the weight of network and the basis function center aiming atthe problems of slow convergence speed of traditional RBF neural networktraining and easily falling into the local minimum. At last the modelingmethod is extended to multiple-input and multiple-output system fields.The model is checked by usage of correlative data, comparison of modeloutput and the actual output, the results show that the improved model is muchcloser to the actual boiler model, which also has a good effect in the multiple-input and multiple-output system modeling, so this improved modeling method issuccessful. The original DCS system of the power station has a lot of problems,such as aging of the system, bad control precision, frequent breakdown, the original system is improved by applying the Siemens PCS7system, and theconfiguration interface is established by WINCC. The operation data could bereal-time monitored in the human machine interface according to the load limit toregulate the related parameters in order to make the working of the boiler securerand more stable.