Online Combustion Optimization System

In this chapter, the online combustion optimization system is exposed. First, the demand of the software system for the practical combustion optimization and the need of the local optimization are summarized, such as data detection requirements, quickness and accuracy requirements, requirements of different optimization goal, requirements online self-learning, parameter optimization limit requirements, fault tolerance requirements, alarm requirements, compatibility of off-line data processing and optimizing and so on. Second the instruments or sensors for online combustion optimization system are introduced. Then, the online SVM algorithm is presented, mainly about derivation of the Incremental Relations, AOSVR Bookkeeping Procedure, efficiently updating the R Matrix, initialization of the Incremental Algorithm and Decremental Algorithm. In addition, there are three main functions with different modules of online combustion optimization system. They are, respectively, online monitoring and alarm function, online optimization and self-learning function, off-line modeling, and optimization function. Finally the application of online combustion optimization system is discussed.