Research And Application Of Closed-loop Combustion Optimization Control Method For Supercritical Boiler

At present,the thermal power units is the main part of China's power supply.The coal burning process of power stations will generate a large amount of polluting gases,which will do harm to the environment.Improving the combustion efficiency of boilers through operation optimization technology while reducing emissions of polluting gases such as NOx is of great significance for further tapping the potential of energy conservation and emission reduction in the energy industry,which will also promotes technological progress in the industry.For the boiler combustion system contains a complex reaction process,the combustion process has the characteristics of strong coupling,large inertia and nonlinear grabbing.Under the background of units generally facing the conditions of flexible coal quality and load,it is difficult to obtain satisfactory control effects with the original control strategy under the steady state and open loop optimization.Therefore,it is necessary to combine artificial intelligence technology and advanced control theory to study a new closed-loop dynamic combustion optimization control method.Based on the object characteristics and operation requirements of the boiler combustion system,this paper studies the applicable dynamic modeling of the boiler combustion system and the combustion optimization prediction control algorithm.The main research contents include:1.A multi-model method(FULP)based on unscented kalman filter-least squares support vector machine is proposed to model the boiler thermal efficiency,reheated steam temperature and NOx emissions.The UKFLSSVM algorithm iteratively updates the kernel parameters ?and model parameters to improve the generalization ability of the support vector machine.With the full use of the fuzzy C-means clustering(FCM)to increase the advantages of the different sample sets and the partial least squares(PLS)to integrate the sub-sample results,a dynamic model of the boiler combustion system is established.Finally,the actual operating data of the unit was collected and tested on a 600 MW unit boiler combustion system.2.On the basis of preliminary optimization operation guidance curve by analyzing historical operation data,non-linear modeling and economic predictive control method are used to perform online secondary optimization to achieve the closed-loop dynamic combustion optimization control.When solving rolling optimization,genetic algorithm(GA)is used as the main optimizer.GA is used to search for the global optimal solution in the entire solution area,and sequential quadratic programming(SQP)is used to optimize the GA results,which significantly improves the GA's functionality.The simulation results show that the algorithm can avoid the local optimization of the solution process,the convergence speed and the solution efficiency is satisfactory,which ensures the practicability and accuracy of the algorithm.3.This study completes the development of boiler combustion system optimization control system.After modifying the DCS system configuration and communication connection,the field application of the boiler combustion optimization system was successfully implemented.According to debugging results,the automatic function of oxygen adjustment was improved and the stable operation of the system was carried out.By comparing the target parameters under the same working conditions,the engineering value of the software is analyzed.