Utility Boiler Combustion Optimization Based On Intolligent Optimization Algorithm

Coal-fired utility boilers not only consume a large amount of coal resources, but also discharge a large number of air pollutants. In issue of improving boiler efficiency and reduce coal consumption and control emissions to mitigate the outstanding problems of high energy consumption and heavy pollution, the boiler combustion optimization method based on intelligent algorithm has its unique advantages compared to traditional method, and which can provide effective guidance for power plant boilers operation of high efficiency and low pollution.By analyzing the operating data of 300MW boiler and 1000MW boiler, support vector regression was used to establish the NOAx emissions SVM model and boiler thermal efficiency SVM model, and the performance of SVM model with RBF kernel was compared with that of SVM model with Sigmoid kernel; Besides, the parameters of the model was optimized by adaptive genetic algorithm(GA) and fruit fly optimization algorithm(FOA) respectively, and the prediction accuracy and generalization ability of each model was compared, the results showed that FOA-SVM model has more advantages.Based on the established SVM model of the boiler NOx emissions and thermal efficiency, combined with adaptive genetic algorithm, NOx emissions can be significantly reduced after optimization when NOx emissions was optimized individually without considering the impact on the efficiency, but which will result in decline of the boiler thermal efficiency at the same time; and also, when the thermal efficiency was optimized individually without considering the impact on NOx emissions, the thermal efficiency can be greatly improved, but at the same time, which will result in increase of NOx emissions.Taking 1000MW unit boiler as example, based on the established SVM model of NOx emissions and thermal efficiency, an multi-objective optimization of combustion method based on the improved multi-objective genetic algorithm (modified NSGA-II) was proposed, both the boiler thermal efficiency and NOx emissions was taken into account to optimize the boiler operating parameters, and obtain the results of the Pareto optimal solution set consisting of a series of feasible solutions, in which many feasible solutions meet the demand of boiler efficiency increase and NOx emissions reduce, which provide reference to achieve combustion optimization purposes of high efficiency and low NOx emission for unit operation personnel.