Modeling And Intelligent Optimal Control Strategy For Ethylene Loss Rate Of Distillation Column

Distillation process is the most widely in petrochemical industry and chemical industry,which is also one of most energy-intensive processes in the petrochemical industry.Ethylene distillation is the final column for the separation of cracking gas.The design and operation of ethylene tower is directly determines the level of ethylene product quality,yield and energy consumption.With the energy prices rising recently,distillation system plays a more important role in energy saving in the chemical industries.Operations optimization of the distillation column can reduce ethylene loss,improve the energy efficiency of the tower,and also have great significance to increase enterprise ' s economic benefit.Ethylene distillation process mechanism and the operation is complicated,it's has many interference factors and the following characteristics:(1)Rectifying columns are the high order object with many plates,and with strong nonlinear and slow response to control operation.(2)Rectifying columns are more easily affected by many factors.So it is difficult to model by mechanism of ethylene distillation column.The important operation index in rectifying column tower kettle of ethylene loss is measured mainly rely on industrial chromatograph or manual analysis,which exists the problems such as high operating costs and time lag.The current researches on the problem of ethylene loss are mainly use data-driven method to create one single model.So the operating mode of rectifying column changes,the model is difficult to meet the demand of precision and the subsequent operations optimization.In this paper,the working principle of the ethylene rectifying column is analyzed.With the actual production and advanced modeling and optimal control theory,the modeling methods of rectifying column ethylene loss rate and operation optimization problems is studied,the main work is as follows:(1)Review the topic research background,ethylene industrial development and the present research on ethylene distillation modeling and optimal control,etc.The meaning of this research and the working principle of ethylene distillation column are briefly discussed.And the influential factors of the ethylene rectifying column loss are analyzed in this paper.(2)Briefly introduce the reaction mechanism of ethylene distillation column.Important parameters of the ethylene loss are analyzed such as tower kettle flow,tower kettle pressure,tower kettle temperature,refrigerant flow,quantity of reflux,etc.In view of that when condition changes,the model accuracy is reduced.A method of modeling for ethylene loss rate of distillation column with respect to operation conditions is proposed.Operating load and coolant temperature are chosen as attributes,and FCM clustering algorithm is used to classify operational data.Principal component analysis is used to select the main parameters affecting the ethylene loss rate.And tower kettle flow,tower kettle pressure,tower kettle temperature and refrigerant flow are closed as the input variable of the model.The ethylene loss rate is the output variable.PSO is used for parameter optimization and LSSVM algorithm is used to model the ethylene loss rate after operation conditions classification.Higher prediction precision and more rationality of the method are illustrated by simulation and analysis.(3)By minimizing the ethylene loss rate of the distillation column as the optimization objective.The particle swarm algorithm is employed to get the optimal cold agent to control the ethylene loss rate.Practical data are used for simulation analysis for simulation analysis.The simulation results show that the average ethylene loss of distillation column is halved and energy efficiency level is improved after optimization after optimization.