بهينه سازي شكل مغناطيس دائم در ماشين هاي سنكرون مغناطيس دائم با روش هاي تحليلي-عددي

چكيده انگليسي :

Magnet Shape Optimization of Permanent Magnet Synchronous Machines by Analytical Numerical Methods Hooshang Mirahki h mirahki@ec iut ac ir Date of Submission 2018 12 19 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree Ph D Language Farsi st1 Supervisor Dr Mehdi Moallem moallem @cc iut ac ir 2nd Supervisor Dr Mohammad Ebrahimi mebrahim@cc iut ac ir1st Advisor Dr Babak Fahimi fahimi@utdallas eduAbstractPermanent Magnet Synchronous PMS machines are attracting a growing interest for a wide range ofindustrial commercial and transportation applications because of their high efficiency high powerdensity low maintenance compact structure and rather simple control Several methods have beenpresented to reduce cogging torque torque ripple and to increase average torque in SPMSM drive These methods are divided into two categories namely a methods based on modifying the controlschemes and or b changing the motor geometry Changing of the geometry of an electrical motorcan be further divided into parameter and or shape optimization methods In parameters optimization the shape of the motor is optimized by changing a set of design parameters while shape optimizationallows us to find alternate shapes for different parts of electric machines Majority of the presentedmethods for shape optimization in permanent magnet motors are based on FEM which is highly timeconsuming or use parameter optimization which cannot find the best shape of the design region In this thesis two analytical methods based on conformal mapping are presented for shapeoptimization of permanent magnets in a Surface Permanent Magnet SPM and Interior PermanentMagnet IPM motors The aim of this optimization is to reduce torque ripple and cogging torquewhile increasing the average torque or keeping it at its aprior value Therefore in SPM motor designregion is divided into finite cells so that each cell can be assigned with either air OFF or magnet ON To validate the method presented for magnet shape optimization results of a random non manufacturable model are compared with finite element method results The objective function isweighted combination of the amplitude of cogging torque torque ripple and average torque simultaneously The optimization process is repeated for two asymmetrical and symmetrical cases The optimization problem is solved using Genetic Algorithms In order to avoid non manufacturingcases a cluster based method is used For comparison between analytical on off method withconventional on off method optimization are applied into a sample case for two methods Thesimulation results indicated the proposed method is very faster than conventional on off methodwhich is based on finite element method In this thesis two analytical methods based on conformal mapping with application of new techniquesand Lumped Parameter Models LPM are presented for field calculation under no load condition inIPM motors In first method a combination of Lumped Parameters Model Quasi Poisson s equationsand Conformal Mapping methods is used for predicting radial and tangential air gap flux density ofIPM machine In second method the analytical method is based on the Schwarz Christofel SC mapping and LPM method Since in this method the shape of magnet flux barrier and magnetizationdirection can be consideration therefore can be used to rotor shape optimization At last the secondmethod is used for rotor shape optimization of IPM machine Keywords Conformal Mapping Topology Optimization Permanent magnet motor Schwarz Christofel mapping 1

ميره كي، هوشنگ

بهينه سازي شكل مغناطيس دائم در ماشين هاي سنكرون مغناطيس دائم با روش هاي تحليلي-عددي

روش نگاشت همديس , بهينه‌سازي توپولوژي , موتورهاي مغناطيس دائم , نگاشت شوارتس كريستوفل