diplomska naloga
Noles Jovanovič Slana (Author), Zorka Novak-Pintarič (Mentor), Zdravko Kravanja (Co-mentor)

Abstract

V diplomskem delu je prikazan dvostopenjski poenostavljen postopek za načrtovanje fleksibilnih omrežij toplotnih prenosnikov (OTP). V prvi stopnji izvedemo sintezo omrežja pri nominalnih pogojih s superstrukturnim pristopom in mešano celoštevilskim nelinearnim programiranjem. Nato v drugi stopnji pri fiksirani optimalni strukturi izvedemo načrtovanje fleksibilnega omrežja s stohastičnimi metodami. V diplomskem delu smo se osredotočili na eno skupino negotovih parametrov, tj. vtočne temperature toplih in hladnih tokov, za katere smo predpostavili, da so slučajne spremenljivke, opisane z normalno porazdelitvijo in se spreminjajo za največ  5 K od nominalne vrednosti. Pristop smo preizkusili na štirih primerih sinteze OTP. Sintezo pri nominalnih pogojih smo izvedli s programom Synheat. Za dobljena omrežja smo izvedli teste fleksibilnosti in izračunali indekse fleksibilnosti, ki so pokazali, da takšna omrežja niso sposobna tolerirati niti manjših sprememb vtočnih temperatur. Za načrtovanje fleksibilnih omrežij smo uporabili stohastični metodi Gaussove integracije in Monte Carlo simulacije ter aproksimacijski metodi z nominalno in bazno točko. Metode smo primerjali glede na težavnost modeliranja, velikost problemov in računalniški napor za reševanje. Rezultati primerov kažejo, da so skupni stroški fleksibilnih omrežij do 18 % višji od stroškov nefleksibilnih omrežij, saj se poveča skupna ploščina prenosnikov. Aproksimacijski metodi z nominalno in bazno točko dajeta slabše rezultate od stohastičnih metod, vrednosti namenske funkcije so višje do 1,7 %. Pri večjih omrežjih je smiselno uporabiti enostavnejše aproksimacijske metode, pri manj obsežnih primerih pa natančnejšo Gaussovo metodo ali simulacijo Monte Carlo. V primeru negotovih parametrov s simetrično normalno porazdelitvijo uporaba bazne točke ne pomeni bistvene izboljšave v primerjavi z nominalno točko.

Keywords

kemijska procesna tehnika;procesna tehnika;omrežje toplotnih prenosnikov;načrtovanje fleksibilnih omrežij;fleksibilna omrežja;

Data

Language: Slovenian
Year of publishing:
Source: Maribor
Typology: 2.11 - Undergraduate Thesis
Organization: UM FKKT - Faculty of Chemistry and Chemical Engineering
Publisher: [N. Jovanovič Slana]
UDC: 66.045.1(043.2)
COBISS: 14251030 Link will open in a new window
Views: 1827
Downloads: 169
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Design of flexible heat exchanger networks
Secondary abstract: This work demonstrates a simplified two stage procedure for the design of flexible heat exchanger networks (HEN). In the first stage we perform synthesis of network at nominal conditions by superstructure approach and a mixed integer non-linear programming. Then in the second stage we perform a design of a flexible network for the fixed optimal structure by stochastic methods. In this diploma work we focused on one set of uncertain parameters. These are inlet temperatures of the hot and cold streams which we presumed that they are random variables described by a normal distribution in the range of  5 K from the nominal value. The approach was tested on four HEN synthesis examples. We performed the synthesis under nominal conditions by the Synheat programme. Flexibility tests and flexibility index of optimal HENs were calculated. They showed that such networks are unable to tolerate even minor changes of the inlet temperatures. We used stochastic Gaussian integration method and Monte Carlo simulation for designing the flexible networks. Besides, the approximation methods with a nominal point and a base point were used. We compared the four methods regarding the difficulty of modelling, the size of the optimization models, and the computer effort to solve. The results of examples show that the total costs of flexible networks are up to 18 % higher than the costs of inflexible systems, because the total area of the exchanger networks increases. Approximation methods with a nominal point and a base point give worse results than the stochastic methods. The value functions are higher up to 1,7 %. For large networks it is more appropriate to use simpler approximation methods, while in less extensive cases the Gaussian or Monte Carlo methods. The use of a base point method does not mean essential improvements in the case of uncertain parameters with symmetrical normal distribution compared to the simpler nominal point method.
Secondary keywords: heat exchanger networks;network design;flexible heat exchanger networks;
URN: URN:SI:UM:
Type (COBISS): Undergraduate thesis
Thesis comment: Univ. v Mariboru, Fak. za kemijo in kemijsko tehnologijo
Pages: XVIII, 86 str.
Keywords (UDC): applied sciences;medicine;technology;uporabne znanosti;medicina;tehnika;chemical technology;chemical and related industries;kemijska tehnologija;kemijske in sorodne industrije;heat treatment operations and equipment;toplotni obdelovalni procesi in oprema;
ID: 1011076