Study Of Key Technologies For Real-time Optimization System Of Petrochemical Plant

The technique of real-time optimization using combination of simulation and control is becoming a new trend in automatization field in plants. This technique can give out the optimal operational conditions of decision variables automatically and continuously with changing external environment and operational targets according to the real-time measurement information. The optimal operational conditions of decision variables can be used as the set value of basic control loop or advanced control system to achieve the aim to reduce energy consumption and maximize economic benefits. Several key techniques in real-time optimization were pointedly studied in this work including data communication interface, data rectification and operation optimization.The first part of this thesis has carried on detailed exposition to the origin and development of real-time optimization technique consulting lots of relative literature material. The components of real-time optimization system were analyzed on this basis. The network framework of this system was then designed; the functions of each component were described. The architecture of the whole system improved manufacture management and ensured operation safety.In the following part of this thesis, the OPC, ActiveX and OLE Automation techniques were used to develop the data interfaces for various functional modules which realized the open data exchange platform with the core of real-time database. This platform performed a good effect in giving speed and stable data support to modules in system and prepared for close-loop optimization. Implementation schemes for different data interfaces were also interpreted detailedly in this part.Data reconciliation modules and gross error detection and identification method were studied in the last part and solving method was also tested. In this part, steady model parameter fitting was fulfilled by using the results of data reconciliation. Furthermore, the optimization algorithms adopted by HYSYS optimizer were analyzed respectively and the results of different algorithms were compared through experiments.This thesis contains many key techniques in real-time optimization. The data exchange platform developed in this work and the theoretical researches in data reconciliation, gross error detection, parameter fitting and process optimization laid a solid foundation for the further development and implementation of real-time optimization technique.