An Application Research On Optimization Algorithm Of Power Plant Coal-Fired Boilers' Combustion Process

Boilers have an important place in the energy industry. At present the large capacity boilers of sub-critical and supercritical are used most in our country coal-fired power stations, these boilers have generally double problems of high emissions of pollutant and low combustion efficiency because of self-defects of equipments and performance management. With the increasing severity of environmental problem and strength of energy saving and emission reduction in China, coal-fired utility boilers operation are confronted with double stress to increase thermal efficiency for reducing their operation cost and to lower their emissions mainly the emission of NOx.On the above-cited situation, experts and scholars from various countries have been committing themselves to the topic and trying to find effective ways to resolve this problem. Based on depth study of the characteristics of the boiler, aim at the complex properties of the boiler combustion system, such as multi-variables, close coupling, multi-disturbances and big lagging. In this paper, a support vector machine was used to set up the combustion characteristic models of the boiler in power station, and parameters of the SVM models were optimized by the mixed arithmetic of the Genetic Arithmetic and Artificial Fish-Swarm Algorithm, then optimized operating was done to working conditions of the combustion characteristic models by using the mixed arithmetic to find the best objects and their corresponding parameters groups of the selective working conditions. Finally we could accomplish the purpose of boiler combustion optimization by regulating these alterable parameters.The content of this thesis mainly includes the following aspects:(1)This paper firstly describes the research significance of the boiler combustion optimization, introduces the present research situation and developing trend of the topic. We establish the paper's starting point base on some problems existing in the present research fruits.(2)Analyses the mechanisms of NOx formation in the process of boiler combustion, puts forward a new method based on reverse equation means to calculate the boiler thermal efficiency, and uses SVM to set up the mathematic models of the optimization objects.(3)In the end, this thesis brings forward a mixed arithmetic of the Genetic Arithmetic and Artificial Fish-Swarm Algorithm then uses it for searching the favorable parameters of the SVM models and the best objects in working conditions of combustion optimization, the results show that this scheme has a satisfying effect.