Multi-time Delay Nonlinear Systems Decentralized Robust Control For Alumina Sodium Carbonate Decomposition And Its Application

In the process of sintering alumina production,the alumina continuous carbonation decomposition process(ACCDP) is a critical procedure bridging the preceding and subsequent procedure.Combination reaction of the sodium aluminate solution produced by upstream desiliconization process and carbon dioxide gas is occurred,and aluminium hydroxide with certain quality index is produced,at the same time qualified mother solution is supplied in the ACCDP.ACCDP is composed of six series decomposition tanks,where complex multi-phase chemical reactions are involved.As a complicated metallurgy process, ACCDP has many characteristics such as multivariable,nonlinear, strong-coupling,large inertia,large delay,uncertainty from the concentration change of sodium aluminate solution and carbon dioxide gas,disturbance from temperature and level of reaction ax,etc,and many fields such as mass transfer,heat transfer and hydromechanics are related. In this dissertation,based on the analysis of ACCDP,nonlinear multi-delays state-space model of ACCDP is established,the equivalence between T-S fuzzy model of ACCDP and nonlinear multi-delays interconnected model is proved,the decentralized robust H_∞state feedback tracking controller and decentralize state observer are designed for T-S fuzzy nonlinear interconnected multi-delays large system,and based on T-S fuzzy mode,the application platform for ACCDP is developed.The major innovation research achievements include:1.Multi-delays interconnected nonlinear model is established.Base on the analysis for operational mechanism and production process of ACCDP,the modeling methods of ACCDP are investigated.Interaction continuous stirred-tank reactor(ICSTR) model for ACCDP is constructed. On the base of dynamic differential equations derived by the material balance principle,multi-delay nonlinear interconnected state-space equation for ACCDP is established,and numerical example is given to illustrate the effectiveness of the proposed method.2.The equivalence between T-S fuzzy model and its nonlinear multi-delays interconnected model for ACCDP is proved.First,T-S fuzzy model of ACCDP is established,and then the local linear dynamic characteristics of nonlinear system at different state-space regions are described by linear model.As a result,the nonlinear dynamic characteristics are described by nonlinear membership function and linear model.3.For multi-delay interconnected nonlinear systems,based on T-S fuzzy model the stability analysis and stability criterion are proposed,the design approaches for decentralized robust H_∞state feedback tracking controller and decentralized state observer are investigated.The simulations with ACCDP model verify the effectiveness of proposed methods.4.For multi-delay interconnected nonlinear large-scale systems,a decentralized output feedback control strategy based on T-S fuzzy model is proposed.The existence conditions of a decentralized output feedback are deduced,and a decentralized output feedback stabilization controller is obtained by iterative linear matrix inequality approach.5.For ACCDP,the control system hardware structure is designed,and control system software is developed.The control software platform is formed based on T-S fuzzy model,and simulations were done on the platform,the results verified that the decentralized robust control methods proposed in this dissertation for multi-delay interconnected nonlinear large-scale systems is effective,and can be used to ACCDP.