Research On Visual Analysis System Of Refining Process Production Indicators Oriented To Operation Optimization

The petrochemical industry is the foundation of the national economy,it is also an important pillar industry to maintain the sustained and stable development of the national economy.At present,China has entered the ranks of major oil refining countries,but,with the vigorous development of the refining industry,there has been a phenomenon of "big but not strong".Compared with developed countries,China’s refining industry generally has such problems as low resource utilization,high energy consumption cost,poor product quality and serious environmental pollution.At the same time,the continuous expansion of production scale,the increasingly stringent environmental indicators and the increasingly fierce market competition brought about by economic globalization urge refineries to further improve their competitiveness.Optimizing the refining process is of great significance to the production and operation of enterprises.Because of the complex chemical reactions and energy exchange in refining process,it is difficult to establish mechanism model.Researchers usually use the combination of mechanism model and evolutionary optimization algorithm to optimize the operation of refining process.These methods are constrained by both the accuracy of mechanism model and the performance of evolutionary optimization algorithm.At the same time,the existing visual analysis software of refinery process is usually only used to display and monitor the production indicators data.Its function is very limited,and it does not have a deeper analysis and guidance ability,so it can’t be used to solve the problems of refinery process operation optimization research.Therefore,this paper develops a visual analysis system for operation optimization of refining process,aiming at analyzing the potential mode and characteristics of production indicators,determing the search scope of parameters in operation optimization,and then assisting researchers to carry out research on operation optimization of refinery process based on evolutionary optimization.The main work of this paper is as follows:(1)This paper reviews the current research status of visual analysis and optimization of refining process operation,summarizes the relevant theories and technologies of visual analysis.On this basis,the difficulties of operation optimization in refining process are analyzed,the objectives of the system-assisted optimization of operation in refining process are determined,and the requirements of the system to be developed are analyzed.(2)According to the requirement analysis of the visual analysis system of production index of refinery process oriented to operation optimization,the visual analysis process oriented to operation optimization was determined.Around this visual analysis process,a series of algorithms and visual design are proposed to solve the visualization problem of high-dimensional data sets,the visualization problem of causal correlation between indicators,the requirement of production index recommendation system and the requirement of defining the search range of parameters for the optimization research of refining process operation.(3)Combined with the algorithm and visual design proposed in this paper,the overall structure design and function design of the system are carried out,and the detailed functions and realization ideas including user login,data navigation of production indicators,exploration of production indicators relationship,selection of production indicators to be analyzed,scalar interval analysis of production indicators are established.Then,the functions mentioned above are developed by using various development technologies,and the expected goals are achieved.The system uses browser/server mode to separate the front and back ends in software implementation,and communicates through RESTful API,which ensures the extensibility of the system.(4)In order to verify the validity of the system,a series of validation of the system function is carried out based on the DCS system export data of Jiujiang Petrochemical Whole Process Optimizing Platform.Finally,the effectiveness of the system is illustrated by its functional effects and data results.As a visual analysis system oriented to operation optimization,the system can assist and guide researchers to conduct operation optimization research in refining process in a better way.