Optimisation of design and operation policies of binary batch distillation with fixed product demand

Optimal design (vapour load, V and number of stages, N) and operation (reflux ratio, R and batch time, t b) of batch distillation have received significant attention in recent years. In these studies, it has been suggested that V should be set at some upper limit (thought to be optimum) and the opti...

Full description

Saved in:
Bibliographic Details
Published inComputers & chemical engineering Vol. 28; no. 11; pp. 2377 - 2390
Main Authors Miladi, M.M., Mujtaba, I.M.
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 15.10.2004
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Optimal design (vapour load, V and number of stages, N) and operation (reflux ratio, R and batch time, t b) of batch distillation have received significant attention in recent years. In these studies, it has been suggested that V should be set at some upper limit (thought to be optimum) and the optimisation task should be focused on finding the optimum values of N, R and t b, which would minimise the capital investment while maximising an economic objective function usually the profit ( P). The major drawback of such optimisation strategy is that there is no constraint on the amount of product (on specification) being produced ( N B = no. of batches in a given time) and the overall profit can only be maximised by producing unlimited amount of products. Unplanned and unlimited production of products are not sustainable and may lead to significant losses in the case of large inventory requirements of any excess products produced. In this work, for the first time, the optimal design and operation task for batch distillation is considered with due regards to the market demands (small to large number of batches) of the products being produced. An optimisation problem formulation is presented for the task which is different than those used in the past under unlimited product demand scenario. Simulated Annealing type algorithm is used for the solution of the optimisation problem. With several examples, it will be shown that fixing V a priori (say V < V opt) will not allow production of N B batches of products with any combination of ( N, R, t b) in a given production time. Also it will be shown that with V fixed at say V > V opt, P will always be < P max for any combination of ( N, R, t b). The comparisons between the results with and without fixing of V a priori show very clear improvement in the annual profit.
ISSN:0098-1354
1873-4375
DOI:10.1016/j.compchemeng.2004.06.001