Research On Boiler Optimization Control System Based On Neural Network

Nowadays, Thermal power still accounts for the vast majority of China’s powergeneration capacity. The development of social economic makes the electricity safetyand reliability requirements in raising. Boiler in power plant as one of the three majorequipment of the power plays a decisive role. The main products for boiler is thesuperheated steam. maintaining constant steam pressure, has a very important effectto productivity and quality of the power plant. So, studying on steam pressure controlloop and improvement, has been one of the main purposes of optimization control ofboiler.In this paper, based on combustion optimization of220t/h boiler control andfieldbus technology research project of a power plant in Daqing, makes the boilersteam pressure control system optimize the existing modeling and simulation.Firstly, introduce and analyze the structure and training principle of radial basisfunction neural network, to make the groundwork for the identification of the boilersystem; Secondly, elaborate the structure of control scheme of steam pressure boiler,boiler thermal system working principle and field control system in detail, Reflectingthe advantages of fieldbus control system. At the same time, modeling of the boilerbased on radial basis function neural network, Approximation fit to the steampressure, furnace negative pressure and the oxygen content in flue gas. Finally,introduce the fuzzy control and giving the control scheme, The control scheme ismainly around the main steam pressure loop unrolling, Tuning the PID controlsystem of the traditional, Establishing the optimization of fuzzy PID control system.Then compare with the PID control system. The results show that the fuzzy PIDcontrol system is better than that of PID control system. Achieve the aim ofoptimization of steam pressure control system and meet the requirement and standardof thermal power plant.