Multivariable Constrained Predictive Control With Applications To AGC Of CFB

In recent years,with the energy consumption and environmental standards becoming more and more stringent,the modeling and control of circulating fluidized bed boiler(CFBB)unit has increasingly become a hotspot and focus because of its significant advantages such as high combustion efficiency,wide fuel adaptability,low combustion pollution,high desulfurization efficiency and low investment cost.However,there are still many problems in the modeling and control of CFB unit.On one hand,taking the combustion system as an example,the modeling method for the design of CFB mostly focuses on the analysis of internal mechanisms which can increase the difficulty of calculation because of the complex model structure resulting from the complicated flow processes and combustion reactions in CFB.Consequently,this method does not suitable for the research of control strategy.On the other hand,because CFB unit is a typical non-linear,large delay,large inertia,strongly coupled thermal object with input and output constraints,the traditional control structure and strategy which does not make full use of CFB unit's characteristics is difficult to obtain satisfactory control performance.In summary,it has a very important theoretical significance and application value to build a simplified first principle model of CFB unit and study on novel control structure and strategy.Firstly,in this paper,based on the study about the characteristics of CFB unit,a simplified first principle model of CFB unit is built which combines with combustion system,steam system,superheaters and turbine system.In particular,the influences of residual carbon and primary air to the burning in furnace are considered in the modeling of combustion system.Simultaneously,Improved Harmony Search(IHS)algorithm is proposed to apply to the parameters identification of principle model which is tested and verified through experimental data.Secondly,a novel AGC control structure is designed which can make full use of the influence of primary air to residual carbon and heat storage in furnace.At the same time,the theory of zone control is introduced to control strategy considered with the actual requirements of the project.Then,multivariable constrained predictive control algorithm based on HIS-MHE is proposed.The result of simulation based on novel AGC control structure verifies the tracking performance,anti-interference ability and robustness of control algorithm.Finally,based on Matlab and Visual C++ platform,the predictive control algorithm is applied to AGC control software.Correspondingly,some modifications to logical configuration are finished.