Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Conductor size selection in low voltage networks under distributed energy resource uncertainty
Waswa, L. S. 2025. Conductor size selection in low voltage networks under distributed energy resource uncertainty. Unpublished doctoral dissertation. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/95da4468-231b-4161-95a9-d3b93925feda
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613966149091328 |
|---|---|
| access_status_str | Open Access |
| author | Waswa, Lewis Sakwa |
| author2 | Bekker, Bernard |
| author_browse | Bekker, Bernard Waswa, Lewis Sakwa |
| author_facet | Bekker, Bernard Waswa, Lewis Sakwa |
| author_sort | Waswa, Lewis Sakwa |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Waswa, L. S. 2025. Conductor size selection in low voltage networks under distributed energy resource uncertainty. Unpublished doctoral dissertation. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/95da4468-231b-4161-95a9-d3b93925feda |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/132299 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:44:31.934Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/132299 Conductor size selection in low voltage networks under distributed energy resource uncertainty Waswa, Lewis Sakwa Bekker, Bernard Chihota, Justice Stellenbosch University. Faculty of Engineering. Dept. of Electrical & Electronic Engineering. Low voltage systems -- Design and construction Electric networks, Active Electric power systems -- Load dispatching Distributed generation of electric power UCTD Waswa, L. S. 2025. Conductor size selection in low voltage networks under distributed energy resource uncertainty. Unpublished doctoral dissertation. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/95da4468-231b-4161-95a9-d3b93925feda Thesis (PhD)--Stellenbosch University, 2025. ENGLISH ABSTRACT: This dissertation proposes a novel conductor size selection (CSS) methodology for active distribution networks (ADNs) with increasing penetration of distributed energy resources (DERs). CSS is a critical process in the planning, design, and operation of low-voltage distribution networks. This is due to the large number of customers connected at this level. Traditionally, this process considered passive customer loads with limited uncertainties. The low level of uncertainties enabled planners to deterministically perform load estimation and by extension simplified the selection process. The primary technical concerns in passive design were the magnitude of thermal loading and voltage drop. Increasing DER penetration on the distribution network introduces operational and planning uncertainties in the inputs of the CSS process resulting in further complications. DER operational uncertainties arise from the diversity in their output and time of use (ToU) characteristics. Conversely, planning uncertainties are associated with the lack of knowledge of the DER location, capacity, and type. These characteristics are closely linked to the customer or network operator’s decisions and can be dictated by aspects like locational constraints, the customer-notified maximum demand, or financial incentives. Planning uncertainties increase the complexity of the ADN-CSS process and lead to a wider range of quality of supply (QoS) parameters such as voltage rise and imbalance that must be considered in addition to the passive technical concerns. This dissertation identifies and addresses two problems associated with ADN-CSS modelling. The first problem is linked to the scoping of the ADN-CSS process while the second relates to the formulation of an uncertainty-based ADN-CSS methodology. The second problem has three parts: (a) the formulation of an uncertainty-based CSS methodology, (b) the improvement of the computational efficiency of this uncertainty-based CSS methodology, and (c) the alignment of the CSS methodology to local and international assessment standards. To address the first problem, a literature review of existing CSS methodologies is conducted. The review evaluates the changing scope of the CSS process inputs, the implication of these changes to the CSS process, and the modifications needed for alignment towards ADN planning and operations. Focus is placed on the advances in ADN-CSS modelling approaches, and the modifications to the traditional approach that are required to cater to the increasing planning and operational uncertainties. To this end, the study formulates a novel Input-Process-Output (IPO) framework that disaggregates the CSS process into modular modelling components. These include input modelling, computational processing, and output analysis. These components are then applied in their varied forms in the subsequent parts of the dissertation. The second problem is addressed based on the findings from the review that demonstrate the necessity to shift from deterministic to uncertainty-based CSS methodologies, to address the increasing level of DER uncertainties. Subsequently, this study formulates and tests a stochastic-probabilistic CSS methodology using the IPO framework. Findings from this formulation indicate that while using a stochastic probabilistic methodology caters to the uncertainties, the process is computationally inefficient. To this end, this study formulates, tests, and validates an analytical probabilistic approach aimed at addressing the computational intractability associated with the stochastic probabilistic methodology. Lastly, this study integrates a multi-interval (MI) network performance analysis into the analytical probabilistic methodology. This improves the methodology, which has originally been formulated based on the single interval (SI) worst-case performance analysis. This modification further aligns this work with international and local best practice that requires the inclusion of both violation duration and violation severity in determining the network performance. In addressing the second problem, several novel contributions emerge. Firstly, the study embeds a DER hosting capacity (HC) assessment framework in the CSS, replacing the general use of load flow analysis. Secondly, the study enhances the ADN-CSS methodology by merging the operational and planning uncertainties into a single unified probability distribution model. Lastly, this study integrates violation severity and violation duration into the analysis of the MI CSS in line with the local and international compatibility standards. In practice, this work is essential for several reasons. It delineates the CSS process for distribution operators and planners, facilitating the development of alternative methodologies to address the CSS issues. The development of an uncertainty-based CSS approach then offers distribution operators a comprehensive framework for executing ADN-CSS across diverse scenarios. Stochastic-probabilistic ADN-CSS is applicable in scenarios that need robust analysis including performance evaluation at each penetration level. The analytical formulation developed in this study offers faster assessments and is thus particularly valuable for distribution network operators with limited computational resources. Lastly, the integration of MI analysis into the analytical CSS process aligns the ADN CSS formulation to the South African NRS-048-2:2007 assessment standard for evaluation and assessing voltages and further proposes a similar criterion by which thermal loading may be assessed. AFRIKAANSE OPSOMMING: Hierdie proefskrif stel 'n nuwe geleiergrootte-seleksie-metodologie (GSM) voor vir aktiewe verspreidingsnetwerke (AVN'e) met toenemende penetrasie van verspreide energiebronne (VEB's). GSM is 'n kritieke proses in die beplanning, ontwerp en bedryf van verspreidingsnetwerke as gevolg van die groot aantal kliënte op hierdie vlak van netwerk. Tradisioneel het hierdie proses passiewe klantladings met beperkte onsekerhede hanteer. Die lae vlak van onsekerhede het beplanners in staat gestel om lasskatting deterministies uit te voer en het dus die seleksieproses vereenvoudig. Die primêre tegniese bekommernisse in passiewe ontwerp was termiese laste en spanningsval. Toenemende VEBpenetrasie op die verspreidingsnetwerk bemoeilik die GSM-proses en stel operasionele en beplanningsonsekerheid in die insette van die GSM-proses in. VEB operasionele onsekerhede ontstaan uit die diversiteit in hul uitset en tyd van gebruik eienskappe. Beplanningsonsekerheid word geassosieer met die gebrek aan kennis van VEB-ligging, kapasiteit en tipe, eienskappe wat nou gekoppel is aan die kliënt of netwerkoperateur se besluite. Dit kan bepaal word deur aspekte soos liggingsbeperkings, die kliënt wat die maksimum aanvraag, of finansiële aansporings. Beplanningsonsekerhede verhoog die kompleksiteit van die AVN-GSM-proses en lei tot 'n wyer reeks kwaliteit van toevoer parameters soos spanningstyging en wanbalans wat bykomend tot die passiewe tegniese uitdagings in ag geneem moet word. Hierdie proefskrif identifiseer en spreek twee probleme aan wat met AVN-GSM-modellering verband hou. Die eerste probleem is gekoppel aan die omvang van die AVN-GSM proses terwyl die tweede verband hou met die formulering van 'n onsekerheid-gebaseerde AVN-GSM metodologie. Die tweede probleem het weer drie dele: (a) die formulering van 'n onsekerheid-gebaseerde GSM-metodologie, (b) die verbetering van die berekeningsdoeltreffendheid van hierdie onsekerheid-gebaseerde GSMmetodologie, en (c) die belyning van die GSM-metodologie met plaaslike en internasionale assesseringstandaarde. Om die eerste probleem aan te spreek, word 'n literatuuroorsig van bestaande GSM-metodologieë gedoen. Die hersiening evalueer die veranderende omvang van die GSM-proses-insette, die implikasie van hierdie veranderinge aan die GSM-proses, en die wysigings wat nodig is vir belyning met AVNbeplanning en -bedrywighede. Fokus word geplaas op die vooruitgang in AVN-GSM modellering benaderings, en die wysigings aan die tradisionele benadering wat nodig is om voorsiening te maak vir die toenemende beplanning en operasionele onsekerhede. Vir hierdie doel formuleer die studie 'n nuwe Inset-Proses-Uitset (IPU)-raamwerk wat die GSM-proses in modulêre modelleringskomponente verdeel. Dit sluit insetmodellering, berekeningsverwerking en uitsetanalise in. Hierdie komponente word dan in hul uiteenlopende vorme in die daaropvolgende dele van die proefskrif toegepas. Die tweede probleem word aangespreek op grond van die bevindinge van die oorsig wat die noodsaaklikheid demonstreer om van deterministiese na onsekerheid-gebaseerde GSM-metodologieë te verskuif, om die toenemende vlak van VEB-onsekerhede aan te spreek. Vervolgens formuleer en toets hierdie studie 'n stogasties-probabilistiese GSM-metodologie deur gebruik te maak van die IPUraamwerk. Bevindinge uit hierdie formulering dui aan dat alhoewel die gebruik van 'n stogastiese waarskynlikheidsmetodologie wel vir die onsekerhede voorsiening maak, die proses rekenaarmatig ondoeltreffend is. Vir hierdie doel formuleer, toets en valideer hierdie studie 'n analitiese waarskynlikheidsbenadering wat daarop gemik is om die berekeningsspoed wat met die stogastiese probabilistiese metodologie geassosieer word, te verbeter. Laastens integreer hierdie studie 'n multi-interval (MI) netwerkprestasie-analise in die analitiese probabilistiese metodologie. Dit verbeter die metodologie, wat oorspronklik geformuleer is op grond van die enkelinterval slegste-geval prestasie-analise. Hierdie wysiging bring hierdie werk verder in lyn met internasionale en plaaslike beste praktyke wat die insluiting van beide oortredingsduur en erns van die oortreding in die bepaling van die netwerkprestasie vereis. By die aanspreek van die tweede probleem kom verskeie nuwe bydraes na vore. Eerstens sluit die studie 'n VEB-gasheerkapasiteit assesseringsraamwerk in die GSM in, wat die algemene gebruik van lasvloeianalise vervang. Tweedens verbeter die studie die AVN-GSM-metodologie deur die operasionele en beplanningsonsekerheid in 'n enkele verenigde waarskynlikheidsverspreidingsmodel saam te voeg. Laastens integreer hierdie studie die erns van die oortreding en die duur van die oortreding in die ontleding van die MI GSM in ooreenstemming met beide plaaslike en internasionale standaarde. In die praktyk is hierdie werk noodsaaklik om verskeie redes. Dit omlyn die GSM-proses vir verspreidingsoperateurs en beplanners, wat die ontwikkeling van alternatiewe metodologieë vergemaklik om die GSM-kwessies aan te spreek. Die ontwikkeling van 'n onsekerheid-gebaseerde GSM-benadering bied dan verspreidingsoperateurs 'n omvattende raamwerk vir die uitvoering van AVN-GSM oor uiteenlopende scenarios. Stogasties-waarskynlike AVN-GSM is van toepassing in scenarios wat robuuste ontleding benodig, insluitend prestasie-evaluering op elke penetrasievlak. Die analitiese formulering wat in hierdie studie ontwikkel is, bied vinniger assesserings en is dus besonder waardevol vir verspreidingsnetwerkoperateurs met beperkte berekeningshulpbronne. Laastens bring die integrasie van MI-analise in die analitiese GSM-proses die AVN GSM-formulering in lyn met die SuidAfrikaanse NRS-048-2:2007 standaard vir evaluering en assessering van spannings, en stel verder 'n soortgelyke kriterium voor waarvolgens termiese lading geassesseer kan word. Doctoral 2025-06-03T07:39:56Z 2025-06-03T07:39:56Z 2025-03 Thesis https://scholar.sun.ac.za/handle/10019.1/132299 en Stellenbosch University ii, 129 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Low voltage systems -- Design and construction Electric networks, Active Electric power systems -- Load dispatching Distributed generation of electric power UCTD Waswa, Lewis Sakwa Conductor size selection in low voltage networks under distributed energy resource uncertainty |
| title | Conductor size selection in low voltage networks under distributed energy resource uncertainty |
| title_full | Conductor size selection in low voltage networks under distributed energy resource uncertainty |
| title_fullStr | Conductor size selection in low voltage networks under distributed energy resource uncertainty |
| title_full_unstemmed | Conductor size selection in low voltage networks under distributed energy resource uncertainty |
| title_short | Conductor size selection in low voltage networks under distributed energy resource uncertainty |
| title_sort | conductor size selection in low voltage networks under distributed energy resource uncertainty |
| topic | Low voltage systems -- Design and construction Electric networks, Active Electric power systems -- Load dispatching Distributed generation of electric power UCTD |
| url | https://scholar.sun.ac.za/handle/10019.1/132299 |
| work_keys_str_mv | AT waswalewissakwa conductorsizeselectioninlowvoltagenetworksunderdistributedenergyresourceuncertainty |