Optimization Study On Combustion Control System Of Thermal Power Station Based On Data Mining Integrated Model

Along with the development of society, the combination of new subjects such as control science, information science and cognition science has offered certain valuable science theory and access in the field of dealing with complicate system and some important problems in social economy. How to use the new theories and technology to advance the industrialization and modernization is the target and research hot point of every science researcher. As the fuel consumption structure is mainly constituted of coal in China, holding the fuel optimizing and economizing course, and advancing the fuel utilization efficiency are the first and foremost part of the middle-long term fuel countermeasure.At present, control system and information management system of plant unit has been prevalent. This article is targeted at how to discover and totally utilize the hidden knowledge in those systems to optimize the process control and operation of plant unit.In the first section, data mining technology is used to discover the knowledge in thermodynamic objects. Combined with the application background of combustion control system optimization, the general process of data mining application is set forth. Certain valuable data preprocess methods are brought out. Ultimately adopting the DM machine algorithm to analyze the burning system process, analysis the physical meanings of the knowledge rules mined by algorithm and validate the correctness of the rules.The second section is focused on how to use the above mined knowledge information to optimize the process control of thermodynamic objects. The paper advances that apply the human-like intelligence based on rules to optimizing of burning system control and utilize the knowledge information discovery of short delay of parameters to intelligently and rapidly restrain the disturb of unit fuel system side. Consequently it is a useful compensation of the traditional feedback control based on great delay parameter (main temperature stress). Through the simulation, this idea is proved to be feasible and effective.At last the article is to further mining the potential value of DM technology, and introduce it into electric power group on large scale. The comprehensive business integrated model of electric power group based on data mining is first brought out. Bring out the model is continuation of the above research work, and it provides the material realizing environment and access. In electric power group, from the application topic of DM in basic devices group and equipments operation, to the utilization and optimization of knowledge information in superior power business operation and decision-making, all is to retain the competitive competence of the electric power group.