Research Of Neural Network Applied To Combustion Optimizion Of The Boiler In Power Plants

With the fast development of national economy, power generation unit capacity in our country is increasing at a speed of 7~8% a year. At present, about 75% of generation sets are still coal-fired unit in our country, most of which are coal burning unit and the tend will continue for a long period. So the request of automatic control system is more and more high.It not only requests the safety of the boiler combustion and the stabilization,but also it is the important thing that energy-saving and beneficiate ,so this is targeted as one of the most key problems by researchers.The power station boiler is one of the three main frames of thermal power stations. Boiler is quite complex and huge in size,and the process happens in it is multi-variables, and the parameters are heavily coupled together. So its control and optimization are targeted as one of the most key problems by researchers.The boilers combustion process is selected for the studying object in the paper, and a combustion optimizion model for the boiler combustion system is set up by use of RBF neural network.In the paper,the boiler load,the first air quantity,the second air quantity,the output temperature of coal milling and so on are selected for the input parameters,and the hot wind opening of coal milling,the blower opening and so on are selected for the output parameters. The boiler system is modeled through its historical data in the boiler combustion process. In this paper, RBF Neural Network is selected as a tool for modeling according to the characteristics of boiler systems. So it enhances the generalizing ability of RBF Neural Networks and improves its applicability.The parameters which are in their best condition have close relationship with heat efficiency of power station boilers, and this is the criteria of optimal target.The boiler combustion system is modeled through corresponding algorithm. Historical data based on the model which is set up in order to find the best setting value of the coal milling hot wind opening, the best setting value of the best blower opening, the best setting value of providing coal volume,finding. The setting value is provided for DCS in order to instruct the combustion optimizion ,and the model reflects the dynamic characteristic of the boiler combustion. The studying is very good for the theory directing of the boiler designing and optimization.In the end of this paper, the software of boiler combustion optimization system is designed and developed.