International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.

6, December 2012

Genetic Algorithm Application for Optimum Design of Three Phases Disk Type PM Motor
1

B. Yousefi
1

A.Noori Shirazi

S. Soleymani

4*

S. Asghar Gholamian

Department of Engineering, Islamic Azad University-Nour Branch, Nour , Iran. Email: borzoyou@yahoo.com 2 Department of Engineering, Islamic Azad University-Nour Branch, Nour , Iran. Email: abdoreza.noori@gmail.com 3 Department of Engineering, Islamic Azad University-Science and Research Branch, Tehran , Iran. Email: ssoleymani@srbiau.ac.ir 4 Babol University of Technology, Faculty of Electrical and Computer Engineering, Babol, Iran * Corresponding author: S. Asghar Gholamian & gholamian@nit.ac.ir

ABSTRACT
Permanent magnet Double-sided disk type PM motors (or axial flux PM - AFPM) are the most widely used types in different applications. In this paper, the equation related to the determination of design and dimensions of double-sided slotted axial flux motor with internal stator will be investigated. This paper presents a new method for optimum design disk type one-stator-two-rotor surface-mounted permanent magnet motor to achieve high power density. At the end of this paper, the results of optimum design dimensions of Double-sided disk type PM motors based on genetic algorithm is presented with high power density or low total volume.

KEYWORDS
Disk type, Power Density, Genetic Algorithm, PM motor

1. INTRODUCTION
Among different Configurations of permanent magnet electrical motors, disk type PM motors are used in industry because of high efficiency and low volume [1, 2 - 5]. Permanent magnet electrical motor Configurations are divided two groups disk type PM Motor (AFPM) and Radial flux PM Motor (RFPM). Flux direction of disk type PM Motor move parallel with machine shaft opposite of flux direction in RFPM [6, 7]. Among the various configurations of AFPM motors, double-sides disk type PM motors, has the most and best application especially in Electrical Vehicles. Disk type PM motors are designed with slotted stators and without slot (or slot-less), that the slot configuration has the high robustness and power density that without one [1], [6]. This paper presents a new method for optimum design disk type one-stator-two-rotor surfacemounted permanent magnet motor to achieve high power density. 3D view of the slotted doublesided disk type PM motor with one-stator-two-rotor is presented in figure1 [1, 5 and 7].
DOI : 10.5121/ijcsa.2012.2601 1

International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.6, December 2012

Figure1. Typical configuration of a slotted one-stator-two-rotor one It should be mentioned that in this article, software MATLAB2011, was used for programming and all designing limits were considered.

1.1. EQUATION DESIGN OF AFPM MOTORS

Disk type PM motors have double-sided, multi-sided structures and single-sided. Single-sided double . Disk type PM motors is cheapest and easiest structure among other types of AFPMs But because AFPMs. of high magnetism in air gap that causes closing two parts (stator and rotor) and low moment problem using this configuration is not usual. onfiguration High magnetism between Rotor and Stator is defused by using second Stator or rotor that set up ond up. This structure called double-sided. Double-sided disk type PM motors used more than other sided. Configuration. When we need to high power and torque with much limitation on outside diameter, increasing the number of disks is a good suggestion. This structure called multi multi-sided disk type PM motor [5]. Double Double-sided disk type PM motor has featured like: low volume, high efficiency, high torque and power density. density sided In Double-sided disk type PM motor, stator armature winding is as classic synchronous motors and permanent magnetic are placed surface of the rotor. In general, if stator leakage inductance and resistance are neglected, the output power for any are electrical machine can be expressed as

Pout =

m f 1+ 3 K e K p K i A Bg (1 2 )( ) Do m1 2 p 2

(1)

Where, is machine efficiency, m is number of phases of the machine, The quantity Kp is termed the electrical power waveform factor, Ki is current waveform, Ke is the EMF factor which incorporates the winding distribution factor Kw and the per unit portion of the total air gap area spanned by the salient poles of the machine (if any), Nph is the number of turn per phase, Bg is the flux density in the air gap, f is the converter frequency, p is the machine pole pairs, is the diameter ratio for AFPM defined as Di /Do, Do is the diameter of the machine outer surface, Di is the diameter of the machine inner surface [1, 7]. The outer surface diameter Do can be written as

International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.6, December 2012

D o = Pout

m f 1+ K e K p K i A B g (1 2 )( ) 2m 1 p 2

1/ 3

(2)

The machine total outer diameter Dtot for the TORUS motor is given by

Dtot = Do + 2W

(3)

Where, W is the overhang of the end winding from the iron stack in the radial direction. For the back-to-back wrapped winding, protrusions exist toward the axis of the machine as well as towards the outsides and can be calculated as
2 A Dg D i D i2 K cu J s Wcu = 2

(4)

Where Dg is the average diameter of the machine, Js is the current density and Kcu is the copper fill factor. It is noted that the slotted topology motor the depth of the stator slot (Lss) for slotted configuration of this motor is Lss=W. The axial length of the machine Le is given by

L e = L s + 2 L r + 2g

(5)

Where Ls is axial length of the stator, Lr is axial length of the rotor and g is the air gap length. The axial length of the stator Ls is

L s = L cs + 2L ss
The axial length of the stator core Lcs can be written as

(6)

L cs =

B g p D o (1 + ) 4p Bcs

(7)

Bcs = the flux density in the core of stator p = the ratio of average air flux density to peak air gap flux density The axial length of rotor Lr becomes

L r = L cr + L PM
Also, the axial length of the rotor core Lcr is

(8)

L cr =

B u D o (1 + ) 8p Bcr

(9)

International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.6, December 2012

Where Bcr is the flux density in the rotor disc core, and Bu is the attainable flux density on the surface of the PM. The PM length LPM can be calculated as
L PM = r Bg K Br f Bg K d K cg

(10)

Where r is the recoil relative permeability of the magnet, Br is the residual flux density of the PM material, Kd is the leakage flux factor, Kc is the Carter factor, Kf =Bgpk/Bg is the peak value corrected factor of air gap flux density in radial direction of the AFPM motor. These factors can be obtained using FEM analysis [1, 5].

2. Genetic algorithm
One of the finest artificial intelligence methods is genetic algorithm (GA). GA is an easy way solution to optimization and search difficult problems. Genetic algorithms are used in engineering, phylogenetics, computational science, mathematics, physics and other fields.In a genetic algorithm, a population of chromosomes or genotype encode candidate solutions to an optimization problem. Traditionally, solutions are represented in binary as strings of 0s and 1s, but other encodings are also possible. The evolution usually starts from a population of randomly generated individuals and happens in generations. In each generation, the fitness of every individual in the population is evaluated, multiple individuals are stochastically selected from the current population (based on their fitness), and modified (recombined and possibly randomly mutated) to form a new population. The new population is then used in the next iteration of the algorithm. Commonly, the algorithm terminates when either a maximum number of generations has been produced, or a satisfactory fitness level has been reached for the population. If the algorithm has terminated due to a maximum number of generations, a satisfactory solution may or may not have been reached [1 and 2].

2.1. Chromosome representation

Each chromosome in proposed GA is an 1 4 array as shown in Figure2. g Nph A Bg

Figure2. Chromosome representation Air gap length = g Number of turn per phase = Nph Electrical loading= A Flux density of air gap = Bg

2.2. Cross over

For instance, considering one point cross over in figure3, there is not the possibility of mutation in the proposed algorithm. In this study, the selection operator is elected TOURNAMENT.
4

International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.6, December 2012

g g' g A

Nph Nph' Nph Bg

A A' A' A'

Bg Bg' Bg' Bg'

Parent1 Parent2 Children1 Children2

Figure3. One point crosses over

2.3 Fitness function

As noted at the beginning, the goal of this article and proposed algorithm will be used to find the optimum value of power density. The motor power density for the total volume can be defined as

Pden =

Pout

(W / m 3 )

(11)

D Le

2 tot

where, Dtot is the total machine outer diameter including the stack outer diameter and the protrusion of the end winding from the iron stack in the radial direction, Le is the total length of the machine including the stack length and the protrusion of the end winding from the iron stack in the axial direction

3. Result of Optimizing Design

The result of designing double-sides slotted disk type PM motor is presented by genetic algorithm. Rated values of the motor and related limitation are presented in table1. Table1: Rated values of the AFPM motor design parameters Voltage Output power Number of poles Number of phases number of slots Slot fill factor Pole arc ratio Slot per Pole per Phase flux density in stator flux density in rotor Efficiency Residual flux density of PM value 50 V 1.5 Hp 4 3 15 0.78 0.7 1 1.3 T 1.3 T 89% 1.15 T

International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.6, December 2012

By electing generation 1400, and after executing program (software MATLAB2011) several times, algorithm approaches to the optimal value. Figure4 show variations of fitness-function (Pden) for the finest chromosome in every generation. The related consequence to the optimal chromosome (after 315 generations) is shown in table2 that the first line shows the finest chromosome. The details dimension of optimal double-sides slotted disk type PM motor is tabulated in Table3.

Figure4. Power density vs. finest chromosome in each generation Table2: Optimal chromosomes Pden
W cm 3

Nph Turn 74 82 74

A m
15930 17990 15410

mm
1 1.06 1.1

Bg T 0.53 0.53 0.46

Chrom.1 Chrom.2 Chrom.3

0.355 0.35 0.348

Table3: Dimensions of optimal AFPM motor design parameters Voltage power Number of poles Number of phases number of slots Power density Outer diameter inner diameter PM length axial length of rotor axial length of stator protrusion of the end winding Slot depth Teeth width value 50 V 1.5 Hp 4 3 15 0.37 151 mm 75 mm 4 mm 15 mm 15.5 mm 17.5 mm 19.7 mm 8 mm
6

International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.6, December 2012

4. CONCLUSION
The main goals of this paper are: Optimal design of double-Sided disk type one-stator-two-rotor Slotted PM Motors with maximum power density. This paper has designed a 1.5 Hp, 50 V slotted disk type PM motor. In this paper, a genetic algorithm was used for optimization of double-Sided disk type slotted PM motors. Result of simulation and optimization show that the maximum power density of the motor is 0.355 watt per m3.

REFERENCES
[1] S. Asghar Gholamian, S. Hemmati, Reza Nasiri, S. A. Saied and S. Esmaeili Jafarabadi;"Optimum Design of Axial Flux Permanent Magnet Slotted TORUS Motor Using Genetic Algorithm", International Journal of Applied Engineering Research, Volume 4 Number 1 (2009). Uler, G.F.; Mohammed, O.A.; Chang-Seop Koh: Design optimization of electrical machines using genetic algorithms. - Magnetics, IEEE Transactions on , Volume: 31 Issue: 3 , 5-7 Jul 1994, Page(s): 2008 2011. S. Asghar Gholamian and A. Yousefi , "Power Density Comparison for Three Phase Non-Slotted" , Australian Journal of Basic and Applied Sciences , 2010. S. Asghar Gholamian; Evaluation of Power Density for Three Phase Double-Sided Axial Flux Slotted PM Generator, International Journal of Computer Science, Engineering and Applications (IJCSEA) Vol.1, No.6, 2011. Jacek F. Gieras, Rong-Jie Wang and Maarten J. Kamper, "Axial Flux Permanent Magnet Brushless Machines",Publisher: Springer; 2 edition, 2008. Aydin, M.; Huang, S.; Lipo, T.A.; Optimum design and 3D finite element analysis of nonslotted and slotted internal rotor type axial flux PM disc Machines, Power Engineering Society Summer Meeting, 2001. IEEE Volume 3, 15-19 July 2001 Page(s):1409 - 1416 vol.3. Aydin, M.; Surong Huang; Lipo, T.A.; Design and 3D electromagnetic field analysis of non-slotted and slotted TORUS type axial flux surface mounted permanent magnet disc machines, Electric Machines and Drives Conference, 2001. IEMDC 2001. IEEE International2001 Page(s): 645 651.

[2]

[3] [4]

[5] [6]

[7]