| Summary: | Thesis (MScEng) -- Stellenbosch University, 2003. === ENGLISH ABSTRACT: Axial Flux Permanent Magnet (AFPM) machines have become attractive because of significant
improvements in permanent magnets over the past decade, improvements in power electronic
devices, and the ever increasing need for more efficient machines in electric vehicle systems. In
comparison with the cylindrical radial flux motor, the AFPM machine is better in a number of
aspects: short frame; compact construction; high efficiency; brush less construction; good starting
torque and high-power density. The common modes of failure and typical operating conditions
of AFPM machines are discussed further. The focus of this research project is a prototype AFPM
machine developed by the Electrical Engineering Department of The University of Stellenbosch.
The machine considered has a power rating of 300 kW and an operating efficiency of 95 % at a
speed of 2300 rpm. This specific machine is used as an example to illustrate the thermal
characteristics of geometrically similar AFPM machines.
The thermal characterization was achieved with the use of two numerical computer models.
Firstly a fluid model was specially developed and experimentally verified. The objective of the
fluid model was to calculate the mass flow rate of air through any geometrically similar AFPM
machine. The fluid model was further used to investigate the effects of different magnet
thickness and axial gaps between the stator and the rotor plates on the mass flow rate of air
through the machine. The fluid model was verified with experimental testing that was done on a
half-scale Perspex model. During the experimental testing the magnet thickness was varied
between 2.5 mm, 5.0 mm, and 7.5 mm along with axial gaps of 6.5 mm, 7.5 mm, 8.5 mm, and
9.5 mm. The fluid model showed a correlation to within 10 % of the experimental mass flow
rates. The results of these tests showed that the magnet thickness and axial gap between the
stator and the rotor plates had no significant effect on the mass flow rate of air. The fluid model
was based on one-dimensional, steady-state, and incompressible flow.
The second numerical computer model was a thermal model. This model was used to calculate
the transient temperature response of the AFPM machine. The model was based on a twodimensional
transient finite difference solution technique. Experimental temperatures taken from
the prototype AFPM machine were used to verify the thermal model. Correlations between the
experimental and theoretical temperatures were within 5.8 % of each other. The thermal model
was used to investigate the effect of geometrical changes on the temperatures in the AFPM
machine. It was found that these geometrical changes had no significant effect on the
temperatures in the AFPM machine. It was also established that increasing the air mass flow rate over about I kg/s had no further effect on lowering the temperatures. The stator was also
identified as being the most critical component as it reached its maximum temperature limit
before any other component. Heat pipes were considered as an alternative thermal management
technique. The location of the heat pipe was limited to the stator. Further simulations were done
to investigate the effect of the heat pipe properties on the amount of heat removed from the
stator.
Recommendations were made concerning the thermal management of the current and possible
future prototype AFPM machines. It was recommended that a further more detailed investigation
into the use of heat pipes be considered. This recommendation is substantiated by the fact that in
this research project only one type of heat pipe was considered and its location was limited to
within the stator. === AFRIKAANSE OPSOMMING: AFPM masjiene het meer aantreklik geword weens betekenisvolle verbeteringe in permanente
magnete gedurende die laaste dekade, verbeteringe in elektroniese toestelle en die vraag na meer
effektiewe masjiene in elekriese voertuigstelsels. Die AFPM masjien is beter as die Silindriese
Radiale Fluksie Motor wat die volgende aspekte betref: die kort raamwerk; kompakte
konstruksie; hoe effektiwiteit; borsellose konstruksie; goeie aanvangsdraaimoment; en hoe-krag
digtheid. Die algemene vorms van faling en ook die tipiese werkstoestande van die AFPM word
verder bespreek. Hierdie navorsingsprojek fokus op die prototipe AFPM masjien wat ontwikkel
is deur die Elektriese Ingenieurs Departement van die Universiteit van Stellenbosch. Die
masjien onder bespreking wek 300 kW per uur op en is 95% effektief teen 'n spoed van 2300
rpm. Hierdie masjien word gebruik om die termiese kenmerke van geometries-gelyksoortige
masjiene te illustreer.
Die termiese eienskappe is bepaal deur die gebruik van twee numeriese rekenaarmodelle.
Eerstens is 'n vloeistofmodel spesiaal ontwerp en eksperimenteel geverifieer. Die doel van die
vloeistofmodel was om die massa vloeitempo van lug deur enige geometries-gelyksoortige
AFPM masjien te bereken. Die vloeistofmodel is verder gebruik om die uitwerking van
verskillende magneetdiktes en aksiale gapings tussen die stator en die rotorplate op die massa
vloeitempo van lug deur die masjien te ondersoek. Die vloeistofmodel is geverifieer deur
eksperimentele toetsing wat gedoen is op 'n halfskaal Perspex model. Tydens die toetsing het
magneetdiktes gewissel tussen 2.5 mm, 5.0 mm en 7.5 mm en die aksiale gapings tussen 6.5 mm,
7.5 mm en 9.5 mm. Die vloeistof model het 'n korrelasie van binne 10 % van die eksperimentele
massa vloeistempo getoon. Die resultate van hierdie toetse het getoon dat die magneetdiktes en
die aksiale gapings tussen die stator en die rotorplate geen noemenswaardige uiterking op die
massa vloeitempo van lug gehad het nie. Die vloeistofmodel is gebaseer op een-dimensionele,
gestadigde, onsamedrukbare vloei.
Die tweede numeriese model was 'n termiese model. Hierdie model is gebruik om die transiente
temperatuur respons van die AFPM masjien te bereken. Die model is gebaseer op 'n tweedimensionele,
transiente eindige-verskil oplossingstegniek.
Eksperimentele temperature gemeet op die prototipe AFPM masjien is gebruik om die termiese
model te verifeer. Die eksperimentele en teoretiese temperature het binne 5.8% met mekaar
gekorrelleer. Die termiese model is gebruik om die uitwerking van geometriese veranderinge op
die temperatuur in die AFPM masjien te ondersoek. Daar is gevind dat hierdie geometriese veranderinge geen noemenswaardige uitwerking op die temperature van die AFPM masjien
gehad het nie. Daar is ook vasgestel dat 'n vermeerdering in die lug massa vloeitempo yerby I
kg/s geen verdere uitwerking het op die verlaging van die temperatuur gehaad het nie. Die stator
is ge-identifiseer as die mees kritiese komponent aangesien dit sy maksimum temperatuur limiet
bereik het voor enige ander komponent, Hittepype is oorweeg as 'n alternatiewe termiese
bestuurstegniek. Die plasing van die pype is tot die stator beperk. Verdere simulasies is
uitgevoer om die uitwerking van die hittepyp eienskappe op die hoeveelheid hitte wat verwyder
word van die stator te ondersoek.
Aanbevelings is gemaak m.b.t die termiese bestuur van die huidige en moontlike toekomstige
prototipes van AFPM masjiene. Daar is aanbeveel dat daar in meer besonderhede ondersoek
ingestel word na die gebruik van hittepype. Die rede hiervoor is dat daar in hierdie studie net
gebruik gemaak is van een tipe hittepyp en dat die plasing daarvan beperk is tot binne die stator.
|
|---|
