Research On Radial Basis Function Network In Coordinated Control System Of 300MW Power Unit

In the thermal power plants,coordinated control system of unit plays a very important role in realizing automation of power grid dispatching and improving economic benefits of thermal power plants.although hydropower and wind power generation have been widely used in recent years,at present more than 70%of them are thermal power generation,which also requires some thermal power plants to meet the demand of power grid load quickly,meanwhile,it needs to ensure the stability of the system.Therefore,how to improve the efficiency of each unit of thermal power plant and ensure the safe operation of the whole system has become an important research topic in thermal power generation.Thermal power unit is a multi-input and multi-output system with complex structure.The traditional coordinated way of boiler following steam turbine has the ability to resp ond to load quickly,but this way will cause the main steam pressure fluctuation in the furnace too large.If the parameters are not selected properly,it will cause a series of chain reactions,so it is necessary to improve the system.In this paper,the development of coordinated control and the problems of nonlinearity and robustness are analyzed.On the basis of material and energy conversion in the unit,the Mathematical model of 300MW unit under different load conditions is established,and the operation conditions of thermal power plant are simulated.The model is representative and fully conforms to the output curve of the thermal power plant compared to its actual conditions.To sum up,a method combining neural network with conventional PID control is proposed,which is applied to coordinated control system to improve the load response rate of the system.Due to the fact that the system has the characteristics of non-linearity,large inertia and large delay,and is a typical non-linear system,however,the neural network has the characteristics of linearizing the non-linear system.Therefore,a multi-input and multi-output model for 300 MW units is established to simulate and verify its dynamic characteristics.After analyzing its dynamic characteristics,a decoupling controller is designed to eliminate the coupling phenomenon of the system itself.Moreover,a unit control method based on RBF network is introduced,after compiled the S function,and compared with the traditional control method.The simulation results show that the method improves the dynamic response of the system obviously,and achieves more stable and faster response.It not only improves the quality of the system,but also has a better anti-disturbance.