Soft-sensing Modeling And Optimal Control Based On The Distillation Yield Rate Of Unit Energy Consumption In Atmospheric Distillation Process

Refining industry is a high energy consumption industry process.Especially,the atmospheric and vacuum distillation process is the most energy-consuming and productive core production process in the refining industry.Therefore,it is of great theoretical and practical significance for petrochemical enterprises to carry out scientific and reasonable energy efficiency evaluation and optimal control strategy in atmospheric and vacuum distillation process in order to achieve energy saving,reduce emission and improve energy efficiency.The text is based on the research background of the National 863 Project "Monitoring,Evaluating and Optimizing Control Technology and System for Energy Efficiency in the Petrochemical Industry".On the basis of the existing three-level energy efficiency indicator system,the energy efficiency indicator of the distillation yield rate of unit energy consumption is established,and the soft measurements are carried out.Finally,the outlet temperature and top temperature of atmospheric tower are optimized with the maximum of the indicator as the optimization objective,which provides a new theory and technology for petrochemical enterprises to improve energy efficiency.The main research work is as follows:(1)Refining process has the characteristics of complex and long production process,diversification of energy and materials in the production process.Combined with the actual working conditions of energy flow and material flow in atmospheric pressure process,this paper synthetically considers the related factors of energy consumption,raw material input and product output in the process,and establishes a new energy efficiency indicator--the distillation yield rate of unit energy consumption,which can evaluate the energy utilization more reasonably.(2)Industrial chromatograph is usually used to measure the output of atmospheric side-line products,but the equipment investment and the operation cost are high.Soft sensor is proposed to monitor the distillation yield rate of unit energy consumption on-line.The working conditions of atmospheric production process are complex,and the prediction errors of existing methods dealing with single working conditions and single model are too large.Therefore,a soft sensor model based on FCM working conditions division is proposed in this paper.According to the process and reaction mechanism,the working conditions of refinery atmospheric pressure process are divided by fuzzy clustering algorithm.Then,PCA algorithm is used to screen the main parameters and the input variables are determined.The LSSVM algorithm optimized by IPSO is used to establish the prediction model for each sub-working conditions.The simulation results of actual production data show that the proposed method has higher prediction accuracy and lays a foundation for optimal control of energy utilization efficiency.(3)The energy efficiency level of atmospheric refining process directly affects the economic benefits of refinery production units.In order to improve the energy efficiency of atmospheric process,this paper presents an optimal control scheme with the maximum the distillation yield rate of unit energy consumption as the optimization objective.The outlet temperature and top temperature of atmospheric tower are selected as the optimization operation variables,and the optimal control model is established.The IPSO algorithm is used to solve the problem,and the optimal outlet temperature and top temperature of atmospheric tower are obtained.The optimization results show that the control strategy effectively improves the energy efficiency of enterprises.