Analytical modelling and design optimisation of the Eddy Current Slip Coupler

Date
2017-03
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: In this study a novel analytical torque calculation method is proposed and used in conjunction with existing design optimisation algorithms to design and manufacture a double sided radial axis eddy current coupler. The eddy current coupler is a rotating damper that is placed between two mechanical systems, it filters out high frequency torque and protects system components from over-torque. The analytical torque calculation method called the Computationally E cient Finite Element Analysis (CE-FEA) method is a hybrid method that consists of a static FEA and analytical calculations which sets it apart form Maxwell's equations, magnetic equivalent circuits and transient FEA. The CE-FEA has been extensively tested in this study and has been found to be accurate within 5 % of the 3D FEA. The benefit of using the CE-FEA besides its accuracy, is its speed. It is able to provide a result in a fraction of the time a 3D FEA takes The CE-FEA is then used in conjunction with multi objective design optimisation algorithms to find an optimal coupler design. Two types of optimisation algorithms are evaluated and compared, a gradient (MMFD) and a population (NSAGA-II) based method. From the results of the population based method a pareto front is produced which is then investigated by adding a variable based colour gradient map. This provides the designer better insight into the effects different variables have on the size, cost, weight and performance of the machine. Further, the optimal eddy current coupler design is chosen and manufactured and tested in a laboratory environment in order to verify the entire design process. The built coupler measurements didn't initially match that of the simulations. This was found to be due to a low grade of aluminium that was used during the manufacturing of the conductor. The resistivity of the conductor was adjusted in the simulations to match that of a lower grade. With this adjustment the measured and simulated results matched, verifying the design process. Finally, two eddy current coupler case studies are conducted. The rst is a validation of the design process but for a single sided eddy current yaw damper for use in wind turbine drivetrains. The second is a brief comparison between the eddy current coupler and an existing Slip Permanent Magnet Coupler (SPMC).
AFRIKAANSE OPSOMMING: Hierdie studie stel voor 'n nuwe analitiese draaimoment berekening metode wat gebruik word saam met huidige ontwerp optimering algoritmes om 'n tweesydige radiale as werwelstroom koppeling te ontwerp en te vervaardig. Die werwelstroom koppeling is 'n roterende demper wat geplaas word tussen twee meganiese stelsels en filtreer hoë frekwensie draaimoment en dien as stelsel beveiliging teen hoë draaimoment impulse. Die analitiese draaimoment berekening methode genoem die Berekening Effektiewe Eindige Element Analise (BE-EEA) is 'n hibriede metode is. Dit bestaan uit 'n statiese eindige element analise en 'n analitiese berekening wat dit afsonderlik plaas van Maxwell se vergelyking, magnetiese ekwivalente stroombaan analise en oorgangs eindige element analise. Die BE-EEA is omvattend getoets in hierdie studie en daar is gevind dat dit akkuraat is binne 5 % van 3D EEA. Die voordeel van die BE-FEA buiten die akuratheid, is die spoed. Dit is in staat om 'n resultaat te lewer 'n fraksie van die tyd wat 'n 3D EEA sou neem. Die BE-EEA word verder gebruik in 'n veelvuldige doelwit ontwerp optimering algoritme om 'n optimale koppeling ontwerp te vind. Twee tipe optimering algoritmes word geëvalueer en vergelyk, 'n gradiënt (MMFD) en 'n populasie (NSGA-II) gebaseerde methode. Vanaf die resultate van die populasie algoritme word 'n 'pareto' front gelewer. Dit word dan verder ontleed deur 'n kleur gradiënt by te voeg wat afhanklik is van die intree veranderlikes. Hierdie verskaf aan die ontwerper meer insig hoe die intree veranderlikes 'n invloed het op die massa, koste, grote en prestasie van die uittree koppeling se ontwerp. Verder word die optimale werwelstroom koppeling ontwerp gekies, vervaardig en getoets in 'n laboratorium om die ontwerp proses te bevestig. Die koppeling wat gebou is het aanvanklik nie met die simulasie resultate ooreengestem nie. Dit is gevind dat dit as gevolg van 'n lae graad aluminium wat gebruik is vir vervaardiging van die geleier. Die resistiviteit van die geleier is toe in die simulasies aangepas om ooreen te stem met die geleier wat vervaardig is. Met hierdie aanpassing stem die resultate wat gemeet is ooreen met die simulasies, wat dus die ontwerp proses bevestig. Laaste is twee werwelstroom koppeling gevalle studies gedoen. Die eerste is weer 'n bevestiging van die ontwerp proses maar vir 'n enkel sydige werwelstroom gier demper wat gebruik is in wind turbine dryf aste. Die tweede gevalle studie is 'n vergelyking tussen die werwelstroom koppeling en 'n reeds bestaande slip permanente magneet koppeling.
Description
Thesis (MEng)--Stellenbosch University, 2017.
Keywords
Eddy currents (Electric), UCTD, Finite element method, Optimisation . . .
Citation