Heat exchanger network synthesis (HENS) is an important process synthesis problem and different tools and methods have been presented to solve this synthesis problem. This is mainly due to its importance in achieving energy savings in industrial processes in a cost-efficient way. The problem is also hard to solve and has been proven NP-hard (Nondeterministic Polynomial-time) and hence it is not known if a computationally efficient (polynomial) algorithm to solve the problem exists. Thus methods that provide good approximate solutions with reasonable computational requirements are useful. The objective of this thesis is to present new HENS approaches that are able to generate good solutions for HENS problems in a computationally efficient way so that all the objectives of HENS are optimized simultaneously. The main approach in accomplishing this objective is by grouping process streams. This is done either on the basis of the fact that in reality the process streams belong to a specific group or these groups are artificially developed. In the latter approach the idea is to decompose the set of binary variables i.e., the variables that define the existence of heat exchanger matches, into two separate problems. In this way the number of different options to connect the streams decreases compared to the situation where no decomposition is present. This causes the solution time to decrease and provides options for solving larger HENS problems. In this work the multiobjective HENS problem is solved either with the traditional weighting method or with an interactive multiobjective optimization method. In the weighting method the weights are the annual costs of the different objectives. In the interactive multiobjective optimization method the Decision Maker (DM) controls the decision-making process by classifying the objectives at each iteration. This multiobjective approach provides the benefit of using interactive multiobjective optimization, so that it is possible to find the solution that best satisfies the DM without too cognitive or computational load, and compared to the traditional approach of using fixed weights for the objectives, all the possible Pareto optimal solutions can be found. Overall the key value of this work is in presenting ways of simplifying a HENS problem.
|Translated title of the contribution||Lämmönsiirtoverkkojen synteesi virtojen ryhmittelyyn perustuvalla monitavoiteoptimoinnilla|
|Publication status||Published - 2012|
|MoE publication type||G5 Doctoral dissertation (article)|