Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
The application of machine learning for computational electromagnetic solver selection
Thesis (MEng)--Stellenbosch University, 2023.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2023
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613868349456384 |
|---|---|
| access_status_str | Open Access |
| author | De la Bat, Willem |
| author2 | Botha, Matthys |
| author_browse | Botha, Matthys De la Bat, Willem |
| author_facet | Botha, Matthys De la Bat, Willem |
| author_sort | De la Bat, Willem |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (MEng)--Stellenbosch University, 2023. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/127374 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA en_ZA |
| last_indexed | 2026-06-10T12:42:59.065Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2023 |
| publishDateRange | 2023 |
| publishDateSort | 2023 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/127374 The application of machine learning for computational electromagnetic solver selection De la Bat, Willem Botha, Matthys Ludick, Danie Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Machine learning Computational electromagnetic Moments method (Statistics) Thesis (MEng)--Stellenbosch University, 2023. ENGLISH ABSTRACT: The field of Computational Electromagnetic (CEM) encompasses a number of methods for resolving the electromagnetic (EM) response of conducting objects with arbitrary shapes and sizes. The specific geometry and configuration of the problem, along with substantial user expertise, are usually required in order to choose the most effective and accurate CEM solver techniques. The process of selecting the optimal techniques can therefore become highly time-consuming. By utilizing Machine Learning (ML) models to predict, prior to simulation, whether a specific CEM technique will be sufficiently accurate or not, we are able to streamline the antenna design process. In this work, this approach was applied to the hybrid Method of Moments (MoM) and Single Reflection Physical Optics (SRPO) technique, the hybrid MoM and Multiple Reflection Physical Optics (MRPO) technique as well as the Domain Green’s Function Method (DGFM) technique. The first two focusing on feed-reflector type antenna problems and the latter on regular and irregular antenna arrays. Using ML algorithms that include the Decision Tree Classifier (DTC), Logistic Regression Classifier (LRC), Artificial Neural Network Classifier (ANNC) and the Random Forest Regressor (RFR), amongst others, it is shown that it is, in fact, possible to make highly accurate predictions of the accuracy of these CEM techniques, prior to simulation. ENGLISH ABSTRACT: The field of Computational Electromagnetic (CEM) encompasses a number of methods for resolving the electromagnetic (EM) response of conducting objects with arbitrary shapes and sizes. The specific geometry and configuration of the problem, along with substantial user expertise, are usually required in order to choose the most effective and accurate CEM solver techniques. The process of selecting the optimal techniques can therefore become highly time-consuming. By utilizing Machine Learning (ML) models to predict, prior to simulation, whether a specific CEM technique will be sufficiently accurate or not, we are able to streamline the antenna design process. In this work, this approach was applied to the hybrid Method of Moments (MoM) and Single Reflection Physical Optics (SRPO) technique, the hybrid MoM and Multiple Reflection Physical Optics (MRPO) technique as well as the Domain Green’s Function Method (DGFM) technique. The first two focusing on feed-reflector type antenna problems and the latter on regular and irregular antenna arrays. Using ML algorithms that include the Decision Tree Classifier (DTC), Logistic Regression Classifier (LRC), Artificial Neural Network Classifier (ANNC) and the Random Forest Regressor (RFR), amongst others, it is shown that it is, in fact, possible to make highly accurate predictions of the accuracy of these CEM techniques, prior to simulation. AFRIKAANS OPSOMMING: Die veld van Rekenkundige Elektromagnetika (REM) sluit ’n aantal metodes in om die elektromagnetiese (EM)-reaksie van geleidende voorwerpe met arbitrˆere vorms en groottes op te los. Die spesifieke geometrie en konfigurasie van die probleem, tesame met aansienlike gebruikerskundigheid, word gewoonlik vereis om die mees doeltreffende en akkurate REM-oplossingstegnieke te kies. Die proses om die optimale tegniek te kies kan dus baie tydrowend raak. Deur Masjienleer (ML)-modelle te gebruik om voor simulasie te voorspel of ’n spesifieke REM tegniek voldoende akkuraatheid sal bereik, hoop hierdie poging om die antenna ontwerp proses te versnel. In hierdie werk is hierdie benadering toegepas op die hibriede Moment Metode (MoM) en die Enkel Refleksie Fisiese Optika (ERFO) tegniek, die hibriede MoM en die Multi Refleksie Fisiese Optika (MRFO) tegniek sowel as die Gebied Green’s Funksie Metode (GGFM) tegniek. Die eerste twee fokus op voer-reflektor tipe antenna probleme en laasgenoemde op gereelde en onre¨elmatige antenna skikkings. Deur gebruik te maak van ML-algoritmes wat onder andere die Besluit-Boom Klassifiseerder (BBK), Logistiese Regressie Klassifiseerder (LRK), Kunsmatige Neurale Netwerk Klassifiseerder (KNNK) en die Lukraak-Woud Regressor (LWR) insluit, word getoon dat dit moontlik is om hoogs akkurate voorspellings van die akkuraatheid van hierdie REM tegnieke te maak, voordat daar gesimuleer word. AFRIKAANS OPSOMMING: Die veld van Rekenkundige Elektromagnetika (REM) sluit ’n aantal metodes in om die elektromagnetiese (EM)-reaksie van geleidende voorwerpe met arbitrˆere vorms en groottes op te los. Die spesifieke geometrie en konfigurasie van die probleem, tesame met aansienlike gebruikerskundigheid, word gewoonlik vereis om die mees doeltreffende en akkurate REM-oplossingstegnieke te kies. Die proses om die optimale tegniek te kies kan dus baie tydrowend raak. Deur Masjienleer (ML)-modelle te gebruik om voor simulasie te voorspel of ’n spesifieke REM tegniek voldoende akkuraatheid sal bereik, hoop hierdie poging om die antenna ontwerp proses te versnel. In hierdie werk is hierdie benadering toegepas op die hibriede Moment Metode (MoM) en die Enkel Refleksie Fisiese Optika (ERFO) tegniek, die hibriede MoM en die Multi Refleksie Fisiese Optika (MRFO) tegniek sowel as die Gebied Green’s Funksie Metode (GGFM) tegniek. Die eerste twee fokus op voer-reflektor tipe antenna probleme en laasgenoemde op gereelde en onre¨elmatige antenna skikkings. Deur gebruik te maak van ML-algoritmes wat onder andere die Besluit-Boom Klassifiseerder (BBK), Logistiese Regressie Klassifiseerder (LRK), Kunsmatige Neurale Netwerk Klassifiseerder (KNNK) en die Lukraak-Woud Regressor (LWR) insluit, word getoon dat dit moontlik is om hoogs akkurate voorspellings van die akkuraatheid van hierdie REM tegnieke te maak, voordat daar gesimuleer word. Masters 2023-03-03T09:09:22Z 2023-05-18T07:18:50Z 2023-03-03T09:09:22Z 2023-05-18T07:18:50Z 2023-03 Thesis http://hdl.handle.net/10019.1/127374 en_ZA en_ZA Stellenbosch University xiii, 110 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Machine learning Computational electromagnetic Moments method (Statistics) De la Bat, Willem The application of machine learning for computational electromagnetic solver selection |
| title | The application of machine learning for computational electromagnetic solver selection |
| title_full | The application of machine learning for computational electromagnetic solver selection |
| title_fullStr | The application of machine learning for computational electromagnetic solver selection |
| title_full_unstemmed | The application of machine learning for computational electromagnetic solver selection |
| title_short | The application of machine learning for computational electromagnetic solver selection |
| title_sort | application of machine learning for computational electromagnetic solver selection |
| topic | Machine learning Computational electromagnetic Moments method (Statistics) |
| url | http://hdl.handle.net/10019.1/127374 |
| work_keys_str_mv | AT delabatwillem theapplicationofmachinelearningforcomputationalelectromagneticsolverselection AT delabatwillem applicationofmachinelearningforcomputationalelectromagneticsolverselection |