论文部分内容阅读
The application of a stochastic simulated annealing based methodology for the synthesis of distillation configurations using (N-1) columns to separate any zeotropic N-component feed into N product streams is described.The distillation sequences considered are described by the number of column sections S,which can be changed from the minimum 2(N-1) to maximum 4N-6.For each specified S,the number of heat exchangers E is then considered and allowed to vary within a range of 2(N-1) to 2,which would generates various types of thermally coupled distillation configurations.The main feature of the methodology is the use of a coding procedure that makes use of a serial of integer numbers to generate and represent various distillation configurations,involving mixed thermally coupled-heat-integrated distillation sequences.In order to ensure the feasibility of the structural search with a random manner,a binary sort tree approach that is originally used in data structure theory is introduced,so that the problem formulation and solution becomes tractable.The total synthesis problem is formulated as an implicit MINLP (Mixed-integer Nonlinear Programming) problem and solved with a simulated annealing algorithm to minimize the annualized cost of system.A five-component separation problem is used to illustrate the systematic approach.The results show that there are thirteen types of distillation schemes can be synthesized by using the suggested methodology,which provides some significant insights for the synthesis and design of multi-component distillation systems.