Predictive Control On Air Input Of Biological Oxidation Pretreatment Process

In the biological oxidation pretreatment process,the air input of each oxidation tank is a key factor affecting the oxidation efficiency between the bacteria and the pulp,and it is very important to ensure the oxygen content in the oxidation tank for the whole production process.However,in the biological oxidation pretreatment process,the system is always influenced by the surrounding environment,so that the system presents non-linearity and hysteresis.The oxidation tank at all levels is widely used as the "more do not less" the principle of input air,resulting in low oxygen utilization,resulting in a lot of energy waste.In view of the above problems,it is imperative to study the intake air volume during bio-oxidation pretreatment.The main work of this paper is as follows:1、For the data containing random noise regression problem,we propose a robust support vector regression algorithm.Through the transformation of the probability inequality,the chance constraint programming is transformed into the second-order cone programming problem,and the mature convex optimization is used to solve.The experimental results show the advantages of the proposed algorithm in dealing with the data uncertainty.At the same time,lay the foundation for improving the accuracy of the air input predictive model.2、In order to improve the utilization rate of oxygen and realize the real-time control of the input air,a nonlinear predictive control model is established.Among them,the online support vector regression is used as the prediction model,and the performance index of the objective function is optimized by the combination of the particle swarm optimization algorithm and the steepest descent principle to realize the real-time control of the input air.The simulation results show that the proposed control model can effectively predict and control the input air in the oxidation tank,and provide a new method for the study of the intake air volume during the biological oxidation pretreatment process.3、When the system is strongly disturbed by the external environment,it is easy tocause the uncertainty of the model.In this case,a robust model predictive control strategy is proposed to ensure real-time control.By introducing the synchronization control of two discrete time chaotic systems,the synchronization of two discrete time systems is ensured when the model is uncertain,and the oxygen supply system is optimized.