Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks
Thesis (PhD)--Stellenbosch University, 2022.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2022
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867614138287521792 |
|---|---|
| access_status_str | Open Access |
| author | Mendes, Angelo |
| author2 | Van Vuurren, J. H. |
| author_browse | Mendes, Angelo Van Vuurren, J. H. |
| author_facet | Van Vuurren, J. H. Mendes, Angelo |
| author_sort | Mendes, Angelo |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (PhD)--Stellenbosch University, 2022. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/124692 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA |
| last_indexed | 2026-06-10T12:47:16.314Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2022 |
| publishDateRange | 2022 |
| publishDateSort | 2022 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/124692 A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks Mendes, Angelo Van Vuurren, J. H. Stellenbosch University. Faculty of Engineering. Dept. of Industrial Engineering. Natural gas compressors Production scheduling Simulated annealing (Mathematics) Flow of gas Compressors UCTD Thesis (PhD)--Stellenbosch University, 2022. ENGLISH SUMMARY: Natural gas transmission networks are the most common means for transmitting natural gas from processing plants to distribution centres. In order to maintain or increase the ow of gas, compressor stations are installed (composed of compressors arranged in parallel or in series). These compressors are typically powered by a portion of the natural gas owing through the pipelines or in some cases by electrical motors. In the literature, it is estimated that 3% to 5% of the gas transported through these networks is consumed by the compressor stations. It is widely agreed that this range of fuel consumption is prohibitively high, given the large amounts of natural gas transported in these transmission networks. Therefore, even small reductions in natural gas (or electricity) consumed by the compressors present in these networks can represent substantial cost savings. These facts make the well-known fuel cost minimisation problem (FCMP) one of the most important problems to solve among the many natural gas operation optimisation problems documented in the literature. In order to maintain natural gas transmission networks, preventive maintenance is routinely performed on the compressors. The durations of these events depend essentially on the tasks to be performed and the availability of human resources. Typically, a major preventive maintenance of a centrifugal compressor may take approximately two weeks. Whenever a compressor is subjected to a major preventive maintenance, it is turned o . Such a compressor is therefore not available to compress the natural gas transported through the transmission network. As in other areas of industry, the starting times of each major preventive maintenance have to be selected strategically in such a way that the operational objectives of the network management can be achieved. From an optimisation point of view, these challenging decisions are modelled mathematically in the well-known maintenance scheduling problem. In this dissertation, a metaheuristic solution approach (the method of simulated annealing) is implemented to minimise compressor fuel cost in a natural gas transmission network through the integration of an annual maintenance plan for compressors in the network. Based on a literature review (on solution approaches that have been adopted towards solving FCMP models and on solution approaches adopted towards solving maintenance scheduling problems), a novel modelling framework is proposed which is capable of integrating compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks. The underlying multi-period mathematical model is an extension of typical single-period FCMP models. Its solution is an operational/maintenance plan based on the best set of values uncovered for the operational and maintenance decision variables which results in the approximate overall minimisation of the fuel consumed by the compressors in the network. The proposed solution approach is applied to three di erent case studies from the literature which were developed for the evaluation of single-period FCMP models. The rst case study involves a linear transmission network, the second a tree transmission network and the third a cyclic transmission network. The time horizon of each case study is 52 weeks. Before execution of the simulated annealing algorithm, a set of experiments are performed in order to determine appropriate combinations of model parameter values for each of the three case studies. The numerical results indicate that the algorithm is capable of finding high-quality solutions to all three instances. AFRIKAANS OPSOMMING: Aardgas-verspreidingsnetwerke is die algemeenste manier om aardgas van verwerkingsaanlegte na verspreidingsentrums te vervoer. Om die vloei van gas te handhaaf of te verhoog, word kompressorstasies ge nstalleer (saamgestel uit kompressors wat in parallel of in serie gerangskik is). Hierdie kompressors word tipies aangedryf deur 'n gedeelte van die aardgas wat deur die pypleidings vloei of in sommige gevalle deur elektriese motors. In die literatuur word daar beraam dat 3% tot 5% van die gas wat deur sulke netwerke vervoer word, deur die kompressorstasies verbruik word. Daar is konsensus dat hierdie hoeveelheid brandstofverbruik uitermate hoog is, gegewe die groot hoeveelhede aardgas wat in hierdie verspreidingsnetwerke vervoer word. Daarom kan selfs klein verminderings in aardgas (of elektrisiteit) wat deur die kompressors in hierdie netwerke verbruik word, aansienlike kostebesparings inhou. Hierdie feite maak die bekende brandstofkoste-minimeringsprobleem (BKMP) een van die belangrikste probleme onder die talle optimeringsprobleme verwant aan aardgas verspreiding wat in die literatuur gedokumenteer is. Ten einde aardgas-verspreidingsnetwerke in stand te hou, word voorkomende instandhouding gereeld op die kompressors uitgevoer. Die tydsduur van so 'n instandhoudingstaak hang hoofsaaklik af van die aard van die taak wat uitgevoer moet word asook van die beskikbaarheid van menslike hulpbronne. Gewoonlik duur 'n groot voorkomende instandhoudingstaak op 'n sentrifugale kompressor ongeveer twee weke. Elke keer as 'n kompressor aan groot voorkomende instandhouding onderwerp word, word dit afgeskakel. So 'n kompressor is dus nie beskikbaar om die aardgas wat deur die verspreidingsnetwerk vervoer word, saam te pers nie. Net soos in ander nywerheidsgebiede, moet die aanvangstye van elke groot voorkomende instandhoudingstaak strategies gekies word, op s o 'n manier dat die operasionele doelwitte van die netwerkbestuur bereik kan word. Vanuit 'n optimeringsoogpunt word hierdie uitdagende besluite in die bekende onderhoudskeduleringsprobleem wiskundig gemodelleer. In hierdie proefskrif word 'n metaheuristiese oplossingsbenadering (die metode van gesimuleerde tempering) ge mplementeer om die brandstofkoste van 'n kompressor in 'n aardgas-verspreidingsnetwerk te minimeer deur die integrasie van 'n jaarlikse instandhoudingsplan vir kompressors in die netwerk. Gebaseer op 'n literatuuroorsig (oor oplossingsbenaderings wat al gebruik is om BKMP-modelle op te los en oplossingsbenaderings vir die oplos van onderhoudskeduleringsprobleme), word 'n nuwe modelleringsraamwerk voorgestel wat kompressor-onderhoudskedulering en brandstofkoste-minimering in aardgas-verspreidingsnetwerke kan integreer. Die onderliggende multi-periode wiskundige model is 'n uitbreiding van tipiese enkel-periode BKMP-modelle. Die oplossing daarvan is 'n operasionele/onderhoudsplan wat gebaseer is op die beste stel veranderlikewaardes wat vir bedryfs- en onderhoudsbesluite gevind word, en wat lei tot benaderde minimering van die brandstof wat deur die kompressors in die netwerk verbruik word. Die voorgestelde oplossingsbenadering word op drie verskillende gevallestudies uit die literatuur toegepas wat ontwikkel is vir die evaluering van enkel-periode BKMP-modelle. Die eerste gevallestudie is gebaseer op 'n line^ere verspreidingsnetwerk, die tweede op 'n boom-verspreidingsnetwerk en die derde op 'n sikliese verspreidingsnetwerk. Die tydhorison van elke gevallestudie is 52 weke. Voordat die gesimuleerde temperingsalgoritme uitgevoer word, word 'n stel eksperimente uitgevoer om geskikte kombinasies van modelparameterwaardes vir elk van die drie gevallestudies te bepaal. Die numeriese resultate dui daarop dat die algoritme in al drie gevalle oplossings van ho e gehalte kan vind. Doctoral 2022-02-03T16:18:57Z 2022-04-29T09:26:49Z 2022-02-03T16:18:57Z 2022-04-29T09:26:49Z 2022-04 Thesis http://hdl.handle.net/10019.1/124692 en_ZA Stellenbosch University xxii, 180 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Natural gas compressors Production scheduling Simulated annealing (Mathematics) Flow of gas Compressors UCTD Mendes, Angelo A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks |
| title | A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks |
| title_full | A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks |
| title_fullStr | A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks |
| title_full_unstemmed | A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks |
| title_short | A modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks |
| title_sort | modelling framework for integrated compressor maintenance scheduling and fuel cost minimisation in natural gas transmission networks |
| topic | Natural gas compressors Production scheduling Simulated annealing (Mathematics) Flow of gas Compressors UCTD |
| url | http://hdl.handle.net/10019.1/124692 |
| work_keys_str_mv | AT mendesangelo amodellingframeworkforintegratedcompressormaintenanceschedulingandfuelcostminimisationinnaturalgastransmissionnetworks AT mendesangelo modellingframeworkforintegratedcompressormaintenanceschedulingandfuelcostminimisationinnaturalgastransmissionnetworks |