Modeling Of Distillation Column And Adaptive Decoupling Control Of Temperature

Rectification column as the primary link in petroleum refining and chemical production,critical to the operation safe and stable chemical production system,has always been the focus of attention of the process control industry.Based on the strong demand of industrial production for energy saving and emission reduction and optimization control,this paper has carried on the thorough theoretical analysis and the research to the process simulation and process control of this high energy consumption device of rectification column.A general dynamic simulation model of the system of rectification column,and according to the dynamic response of the coupling problem existed in the control system are further discussed,and designed corresponding control scheme.The basic contents are as follows.First,the rectification principle of the rectification column is briefly introduced.Combined with the control target of the rectification column of the static characteristics of the distillation process was analyzed according to the correlation between the variables,to determine the amount of basic control and controlled volume.For the further study on distillation process dynamic response process,according to the material and energy balance,to balance the level based on the assumption that,the mathematical model of the tray model,the tower condenser and the bottom reboiler was established by calculating the thermodynamic parameters such as the bubble temperature and the enthalpy.and combined with the phase equilibrium relationship and the molar normalization equation and other constraints,to achieve the distillation process model.Programmed with MATLAB simulation results show that the model can achieve a good dynamic response with different components of each state variable,and the change of steady-state results agree well with the actual situation,it is of great significance for the optimization of the design and the analysis of the control system.The rectification column is a typical multi-input multi-output multivariable system.Through the simulation analysis of the dynamic model,it can be found that the system has the characteristics of non-linearity,hysteresis and time-varying,and under the traditional control,there are serious coupling characteristics of the temperature control system at both ends of the rectification column.According to the control mode of the coupling the feedforward decoupling control of the situation in the accurate model of the controller has better control effect,but the model mismatch condition is very difficult to achieve ideal control effect of decoupling.Aiming at the shortcomings of traditional feedforward decoupling control,this paper proposes a decoupling control based on neural network PID algorithm based on improved particle swarm.the use of neural network super adaptive and self-learning ability to arbitrarily approximate any complex nonlinear system,with the general problem of generalization ability,to solve the object of the precise mathematical model is too dependent on the problem.at the same time,the initial weights of the neural network are optimized by using the improved optimization ability of the improved particle swarm optimization algorithm,to improve the learning rate of the neural network and realize the fast convergence of the control system to achieve the generalized decoupling effect.Based on the above research results,we use a company's fuel oil atmospheric distillation unit training as a means to experiment,to conduct on-site drive device to practical steady-state operation,and has been running the state under different controllers.The results show that the proposed neural network PID controller based on improved particle swarm can be used in industrial control,to achieve effective decoupling temperature control of distillation column ends,improve the stability of the system,which has a certain practical significance for the study on the problem of coupled multi variable system...