The Study On The Synthesis And The Solution Strategy Of The Process System Engineering

Process system synthesis is one of the most important parts of process system engineering, and has been a very active area of research over the recent decades. Because of the complexity and the lack of effective mathematics algorithms for large scale of Mixed Integer Non-Linear Programming (MINLP) problems, the optimization technique of process synthesis is still considered to be the most difficult and immature field. This paper attempts to carry out the methodological research of the nonlinear process synthesis, including the strategy of mathematical methods based on superstructures, effective solution algorithms for complex MINLP problems, reactor networks and synthesis approaches of heat exchanger networks. The research contents of this paper are as follow:1. The first section of paper generally summarized and analyzed to the development and the former study of the process system synthesis. It is still considered to be the most difficult and immature field though development and improvement have been done. It is pointed out that the different methods of thermodynamics analytical method, mathematics programming and artificial intelligence are the present development heading. The development trend will integrate and carry out these methods. With the development and improvement of numeration mathematics and computer technology, the methods of the process system synthesis included the mathematics programming and the artificial intelligence will be more and more important and significant. So it is very important to study and improve the solution strategy and algorithms on MINLP problems2. The paper, which based on summarized using previous literature, lucubrated on Standard Genetic Algorithms and entirely improved in the process of coding and decoding, initial seed generation, selector operator, crossover operator and convergence condition of Genetic Algorithms and utilized Date Mining technology to improve Genetic Algorithms to jump out local trap and enhance calculation efficiency. The new method was proposed on MINLP problems in order to be widely applied in the processsystem synthesis by Genetic Algorithms.3. According to the optimization design and reactor networks design of chemical system engineering, the method which restriction nonlinear programming problems can be transformed to the no-restriction no-linear programming problems was put forward. The Modified Genetic Algorithms proposed in the paper was utilized to solve these problems and the results showed that it was practical and efficient and global optimized solution was got.4. The synchronization optimization model and solution strategy of heat exchanger networks (HEN) were proposed based on modified Genetic Algorithms aim at the present synthesis methods location of HEN. The tests and application in the practice showed that the model and method were practical and efficient. The several fees of the heat exchanger networks synchronization optimization and synthesis can be realized by means of the method.The paper studied on the optimization synthesis method and solution strategy of non-linear process system including the Genetic Algorithms of a large scale mixed integer non-linear programming and reactor networks and exchanger heat networks. And the research lays a good foundation for large-scale non-linear process problems and application in the practice.