MEV Journal of Mechatronics, Electrical Power, an d Vehicular Technology 10 (2019) 17–23 Journal of Mechatronics, Electrical Power, and Vehicular Technology e-ISSN: 2088-6985 p-ISSN: 2087-3379 www.mevjournal.com doi: https://dx.doi.org/10.14203/j.mev.2019.v10.17-23 2088-6985 / 2087-3379 ©2019 Research Centre for Electrical Power an d Mechatronics - Indonesian Institute of Sciences (RCEPM LIPI). This is an open access article under the CC BY-NC-SA license (https://creativecommons.org/licenses/by-nc-sa/4.0/). Accreditation Number: (LIPI) 633/AU/P2MI-LIPI/03/2015 and (RISTEKDIKTI) 1/E/KPT/2015. Load characteristic analysis of a double-side internal coreless stator axial flux PMG Ketut Wirtayasa a, b, *, Pudji Irasari a, Muhammad Kasim a, c, Puji Widiyanto a, Muhammad Fathul Hikmawan a a Research Centre for Electrical Power an d Mechatronics, In donesian Institute of Sciences Jl. Cisitu No. 154D, Bandung, 40135, In donesia b Department of Electrical Engineering, National Taiwan University of Science an d Technology, No. 43, Section 4, Keelun g Rd, Da’an District, Taipei City, 106 , Taiwan c School of Electrical Engineering an d Telecommunications, University of New South Wales 330 Anzac Parade, Ken sington NSW 2033, Australia Received 15 March 2019; accepted 29 November; Published online 17 December 2019 Abstract The main issue of using a permanent magnet in electric machines is the presence of cogging torque. Several methods have been introduced to eliminate it, one of which is by employing a coreless stator. In this paper, the load characteristic analysis of the double-side internal coreless stator axial flux permanent magnet generator with the specification of 1 kW, 220 V, 50 Hz, 300 rpm and 1 phase is discussed. The purpose is to learn the effect of the load to the generator performance, particularly the output power, efficiency and voltage regulation. The design and analysis are conducted analytically and numerically with two types of simulated loads, pure resistive and resistive-inductive in series. Each type of load provides power factor 1 and 0.85 respectively. The simulation results show that when loaded with resistive load, the generator gives a better performance at the output power (1,241 W) and efficiency (91 %), whereas a better voltage regulator (5.86 %) is achieved when it is loaded with impedance. Since the difference in the value of each parameter being compared is relatively small, it can be concluded that the generator represents good performance in both loads. ©2019 Research Centre for Electrical Power and Mechatronics - Indonesian Institute of Sciences. This is an open access article under the CC BY-NC-SA license (https://creativecommons.org/licenses/by-nc-sa/4.0/). Keywords: coreless stator; axial flux permanent magnet generator; load characteristics; resistive load; resistive -inductive in series. I. Introduction Axial-flux permanent magnet generators (AFPMG) offer several benefits, among others, can be made in various alternative topologies [1] and have high power density [2][3]. Their application is not only in the electricity generation sector but also in electric vehicles, industrial equipment [4], aircraft, compact engine generator, and battery charging [5]. The stator of AFPMG can be built with or without iron core. The latter gives some more advantages since it is lighter than the construction of using core, eliminates cogging torque, easy to manufacture, because it does not need lamination and eliminates magnetic forces to the rotor disc [6] as well as having high efficiency [5][7]. Several types of research on the AFPM coreless stator have been conducted and most of them are used in wind turbine applications. Reference [6] analyzes a double-sided coreless-stator 24 poles and 18 coils AFPMG. The best generator performance can be obtained by varying stator thickness and diameter. The highest efficiency is 91.8 % acquired from the combination of the stator thickness and diameter of 8 mm and 150 mm. Design and analysis of three rotors and double stators coreless AFPMG are observed in [8]. By configuring 12 poles in each rotor and 9 coils in each stator, the generator can produce 1.8 kW and 120 V at 500 rpm. Three rotors are used instead of 4 so that reducing the iron loss and the generator weight. A similar pole configuration is found in [9], which is 12 permanent magnet at each of the rotor core (double rotors) and 9 coils in the stator (single stator). The simulation results indicate that the 500 rpm coreless AFPMG can generate nearly sinusoidal voltage and * Correspon ding Author. Tel: +62-81223114327 E-mail address: ktwirtayasa@yahoo.co.id https://dx.doi.org/10.14203/j.mev.2019.v10.17-23 http://u.lipi.go.id/1436264155 http://u.lipi.go.id/1434164106 http://mevjournal.com/index.php/mev/index https://dx.doi.org/10.14203/j.mev.2019.v10.17-23 https://creativecommons.org/licenses/by-nc-sa/4.0/ https://crossmark.crossref.org/dialog/?doi=10.14203/j.mev.2019.v10.17-23&domain=pdf https://creativecommons.org/licenses/by-nc-sa/4.0/ K. Wirtayasa et al. / Journal of Mechatronics, Electrical Power, an d Vehicular Technology 10 (2019) 17–23 18 current waveforms. The amplitude of the waveforms is 100 V and 5 A respectively. In reference [10], design and prototyping of 3 phase, coreless AFPMG with two rotors and one stator is investigated. The configuration of 20 poles on the rotors and 18 coils on the stator is employed. The measurement test at 300 rpm yields terminal voltage, output power, and efficiency, respectively are 200 V, 200 W, and 94.2 %. The paper discussed the load characteristics of a 220 V, 1 kW, 50 Hz, 300 rpm 1 phase coreless axial flux generator. The simulation is conducted analytically and numerically by employing pure resistive load as well as resistive-inductive loads in series. The aim of this research is studying the effect of the load, mainly on the generator output power, efficiency, and voltage regulation. In addition to the load characteristics, the magnetic flux distribution and air gap flux density simulated using FEMM 4.2 software will also be presented. II. Materials and Methods A. The design feature of the machine The generator topology, dimensions, and main parameters are illustrated in Figure 1 and Table 1 respectively. The rotor is the rotating part of a generator where the permanent magnets are arranged on the inside (Fig. 1a). The stator is the stationary part and the place for the winding (Fig. 1b). The stator and rotor are integrated through a shaft to produce electricity. The constructions of the studied double rotor single coreless stator, as well as its dimensions in the axial direction, are shown in Fig. 1c and Fig. 1d respectively. B. The magnetic field in coreless AFPMG The flux paths of the double-sided rotors internal coreless stator AFPMG is depicted in Fig. 2. The stator is made without core (coreless) and the rotor is made of carbon steel. The flux leave north pole (permanent magnet 1) across stator and air gap to the south pole (permanent magnet 2) and then splits into two equal sections, one of them travels toward south poles of permanent magnet 3, and then passing through the stator as well as air gap to the south pole of permanent magnet 4, as shown by arrow signs. NdFeB has been used as the permanent magnet with Br = 1.030 T and the coercive field strength Hc = 796 kA/m. moreover, Air gap flux density (Bmg) and magnetic flux (f) are stated at equation (1) and equation (2) [11][12], Table 1. The dimension of the generator parts Parameter Unit Outer rotor disc radius, Rro 200 (mm) Inner rotor disc radius, Rri 115 (mm) Winding thickness, tw 4 (mm) Rotor yoke thickness, Ly 60 (mm) Number of turns, N1 340 turn Number of poles, 2P 20 poles Number of parallel con ductor, aw 2 Wire diameter, dw 0.8 (mm) Shaft radius, Rsh 30 (mm) Permanent magnet axial height, hm 40 (mm) Inner permanent magnet arc, Wpi 28.9 (mm) Outer permanent magnet arc, Wpo 50.27 (mm) Permanent magnet length, Lp 85 (mm) Air gap length , lg 3 (mm) (a) (b) (c) (d) Figure 1. Generator dimen sions; (a) rotor; (b) stator and its windin g configuration; (c) three dimen sional coreless AFPMG; (d) front view K. Wirtayasa et al. / Journal of Mechatronics, Electrical Power, an d Vehicular Technology 10 (2019) 17–23 19 𝑩𝐦𝐠 = 𝑩𝐫 𝟏+[ 𝝁𝐫𝐫𝐞𝐜(𝒈+𝟎.𝟓𝒕𝐰) 𝒉𝐦 ]𝒌𝐬𝐚𝐭 (1) 𝜙f = 𝛼i𝐵mg 𝜋 8𝑝 (𝐷ro 2 −𝐷ri 2 ) (2) where Br is the remanence flux (T), rrec is the relative permeability of permanent magnet, g is the axial length of the air gap (mm), tw is the winding thickness (mm), hm is the axial height of the permanent magnet (mm), ksat is the saturation factor, αi is the ratio of pole face width to the pole pitch at average radius, Dro and Dri are the outer and inner diameter of rotor disc (mm), and p is the number of pole pairs. C. Single phase equivalent circuit The equation to identify the number of stator turn per phase (N1) and the voltage induced by the rotor when it rotation (Ef) is given by [11] 𝑁1 = 𝜀 𝑉1 𝜋√2𝑓𝑘w1𝜙f (3) and 𝐸f = √2𝑓𝑁1 𝑘w1𝜙f (4) where f is the frequency = 50 Hz, V1 is the terminal voltage of generator (V),  > 1 for generating mode, kw1 is the winding factor at fundamental harmonic. Fig. 3 is the equivalent circuit of AFPMG. When the generator runs and connected to a load, the induced current (Ia) starts flowing in the stator winding, generates magnetomotive force and interacts with the main field produced by the rotor causing a change in direction and magnitude of the magnetic flux in the air gap. This is usually called an armature reaction. The armature reaction voltage lags the current by 90° and is presented by (-jIadXsd) + (-jIaqXsq). The current Iad produces maximum air gap field aligned with the rotor pole (d-axis), and Iaq aligned with the q-axis (between poles). The stator coil has resistance R1 and leakage reactance X1. The value of R1 is found with equation (5) 𝑅1 = 𝑁1𝑙1𝑎v 𝑎p 𝑎w𝜎 𝑠a (5) with l1av is the average length of the stator turn (m), ap is the number of the parallel current paths, aw is the number of parallel conductors,  is the electric conductivity of armature conductor (S/m), and Sa is the conductor cross section (m2). The sum of the armature or mutual reactance Xa and X1 yields synchronous reactance Xs, stated with 𝑋sd = 𝑋ad + 𝑋1 (6) 𝑋sq = 𝑋aq + 𝑋1 (7) where d and q represent the d- and q-axis respectively. For coreless stator, the leakage reactance is assumed close to 0, so that Xsd ≈ Xad, and Xsq ≈ Xaq. Furthermore, 𝑋ad = 2𝑚1 𝜇0𝑓( 𝑁1𝑘w1 𝑝 )2 (𝑅ro 2−𝑅ri 2) 𝑔𝑑 ′ 𝑘fd (8) 𝑋aq = 2𝑚1𝜇0 𝑓( 𝑁1𝑘w1 𝑝 )2 (𝑅ro 2−𝑅ri 2) 𝑔𝑞 ′ 𝑘fq (9) where m1 is the phase number, 0 is the permeability of vacuum, g'd and g'q is the d- and q-axis equivalent air gap length respectively, kfd and kfq is the form factor in the d- and q-axis consecutively, with kfd and kfq equal to 1 for surface configuration of a permanent magnet. The armature current is 𝐼a = 𝐼ad + 𝐼aq (10) If the generator is connected to an electrical load, then Figure 2. The geometry and magnetic flux path of double-sided rotors internal coreless stator AFPMG Figure 3.Single phase equivalent circuit of AFPMG [11] Table 2. The resistive and inductive load Cos 𝛟 = 1 Cos 𝛟 = 0.85 RL() XL () RL() XL () 400 0 400 247.895 360 0 360 223.105 320 0 320 198.316 280 0 280 173.526 240 0 240 148.737 200 0 200 123.947 160 0 160 99.158 120 0 120 74.368 80 0 80 49.579 40 0 40 24.789 30 18.592 K. Wirtayasa et al. / Journal of Mechatronics, Electrical Power, an d Vehicular Technology 10 (2019) 17–23 20 𝐼ad = 𝐸f (𝑋sq + 𝑋L ) (𝑋sd + 𝑋L )(𝑋sq+ 𝑋L )+(𝑅1+ 𝑅L ) 2 (11) 𝐼aq = 𝐸f (𝑅1+ 𝑅L ) (𝑋sd + 𝑋L )(𝑋sq + 𝑋L )+(𝑅1+ 𝑅L ) 2 (12) where RL and XL are the load resistance and reactance in  consecutively. D. Output power and voltage regulation The terminal voltage (𝑉1) and output power (Pout) due to the load impedance (ZL) are calculated with 𝑉1 = 𝐼a𝑍L (13) 𝑃out = 𝑚1 𝑉1𝐼a cos 𝜙 (14) 𝜙 = arccos ( 𝐼a𝑅l 𝑉1 ) = arccos ( 𝑅l 𝑍L ) (15) where ZL is the load impedance and 𝜙 is the power factor angle. The simulated loads RL and XL that give two different load power factor (PF) 1 and 0.85 are listed in Table 2. The percentage change in the output voltage from no-load (Vnl) to full-load (Vfl) when the generator is loaded by unity, lagging and leading power factor, or also called as voltage regulation VR is obtained using equation (16) [13], 𝑉𝑅 = 𝑉nl− 𝑉fl 𝑉fl 𝑥 100 % (16) Generator losses including winding loss P1 (W), eddy current loss Pe (W) and rotational loss Prot (W) are presented by, Ʌ𝑃1 = 𝑚1 𝐼a 2 𝑅1 (17) Ʌ𝑃e = 𝜋2 4 𝜎 𝜌 𝑓 2𝑑2𝑚con [ 𝐵mx1 2 + 𝐵mz1 2 ]𝜂d 2 (18) ⧍𝑃rot = ⧍𝑃fr + ⧍𝑃wind (19) where is the specific mass density of the conductor (kg/m3), mcon is mass of the stator conductor without end connection and insulation (kg), d is the diameter of the stator conductor (m), Bmx1 and Bmz1 are the peak values of tangential and axial components of the magnetic flux density (T) respectively, and ηd is the coefficient of distortion. For the last, the efficiency of the generator is expressed in equation (20) η = Pout Pout + ∆P = Pout Pout+ (∆P1+ ∆Pe +∆Prot) (20) III. Results and Discussions A. Magnetic field distribution The magnetic field distributions of the generator in no-load and on loaded conditions are shown in Fig. 4. For the simulation, the load current correlated with the PF=1 is 5.57 A and for the PF = 0.85 is 6.32 A. At no-load condition (Fig. 4a), the magnetic flux is only produced by permanent magnets on the rotor. The maximum value of flux density B (in the box) is the highest (1.049 T) compared to the underloaded condition. When load RL and RL + JXL are applied, the magnetic flux generated by the current flowing in the conductors suppresses the magnetic flux produced by the magnets, which results in a decrease in the maximum B. Both simulated loads give the same maximum values of B, which is 1.047 T (Fig. 4b & c). In general, all the results in Fig. 4 exhibits good magnetic flux distribution on the stator and rotor indicated by the absence of the flux concentration spots. Besides, all the maximum flux densities are lower than the saturation point 2.2 T. As previously explained, the armature reaction takes place in the air gap. From Figure 5, it can be seen the inverse correlation between the air gap flux density Bmg and the load current. The peaks of Bmg waves can be observed clearly between the no-load and loaded condition but it appears to coincide between the loaded ones due to a very small difference in value. The peak values of each wave are 0.80896 T, 0.7731 T and 0.76785 T for no-load, PF = 1 and PF = 0.85 consecutively. B. Generator performance prediction The calculation results of Ia, V1, Pout, VR and η, at load RL and RL + JXL are presented in Table 3 and Table 4. For easy comparison, the parameters in Table 3 and Table 4 are graphically illustrated as shown in Fig. 6 to Fig 9. In a synchronous generator using a stator core of any size, the winding resistance is frequently neglected because its value is considered too small compared to the synchronous reactance. However, it is different from a coreless generator. From the calculation, it is obtained Xsq = 0.036 and R1 = 2.39 , or in other words, the internal load is more resistive. With two types of the given loads, Ia is higher when the load is RL and this causes a higher internal voltage drop, which finally results in lower V1 Table 3. Calculation results of the electrical parameters at load RL RL () Ia (A) V1 (V) Pout (W) VR ( %)  (%) 400 < 0° 0.59 236.16 137.78 0.60 74.26 360 < 0° 0.65 234.76 152.88 0.66 76.13 320 < 0° 0.73 234.60 171.71 0.75 78.08 280 < 0° 0.84 234.41 195.82 0.85 80.12 240 < 0° 0.97 234.16 227.82 1.00 82.24 200 < 0° 1.17 233.83 272.31 1.19 84.44 160 < 0° 1.45 233.37 338.39 1.49 86.69 120 < 0° 1.93 232.68 446.79 1.99 88.89 80 < 0° 2.87 231.55 657.29 2.99 90.81 40 < 0° 5.57 229.31 1241.53 5.97 91.11 K. Wirtayasa et al. / Journal of Mechatronics, Electrical Power, an d Vehicular Technology 10 (2019) 17–23 21 (Fig. 6). For having better power factor, the generator with load RL produces better output power (Fig. 7) and its efficiency is also slightly higher accordingly (Fig. 8), with the best value of 91.11 % at 5.57 A. Generator with load ZL provides the highest efficiency of 90 % at 4.81 A and then it goes down for saturation. According to Eq. (16), VR represents the ratio of voltage drop (from no load to full load) to the no-load voltage. Therefore, it should be as low as possible to gain a stable power distribution. It is already mentioned that a higher voltage drop occurs when the load is RL (with referring to Fig. 6). Consequently, Table 4. Calculation results of the electrical parameters at load Z L ZL () Ia (A) V1 (V) Pout (W) VR ( %)  (%) 400 < 31.79° 0.50 236.16 99.86 0.44 67.76 360 < 31.79° 0.55 235.13 110.85 0.48 69.93 320 < 31.79° 0.62 235.02 124.56 0.55 72.25 280 < 31.79° 0.71 234.88 142.13 0.62 74.70 240 < 31.79° 0.83 234.70 165.48 0.73 77.31 200 < 31.79° 0.99 234.45 198.00 0.87 80.07 160 < 31.79° 1.24 234.12 246.43 1.09 82.96 120 < 31.79° 1.65 233.61 326.22 1.46 85.93 80 < 31.79° 2.46 232.77 482.36 2.19 88.72 40 < 31.79° 4.81 231.11 924.45 4.39 90.05 30 < 31.79° 6.32 226.23 1198.47 5.86 89.38 (a) (b) (c) Figure 4. Magnetic field distributions under; (a) no-load condition; (b) load R L; (c) load RL + JXL K. Wirtayasa et al. / Journal of Mechatronics, Electrical Power, an d Vehicular Technology 10 (2019) 17–23 22 the VR is also higher with the maximum value of 5.97 %, and for PF = 0.85 or load ZL, VR = 5.86 % (Fig. 9). These values meet the requirement, which is below 8 %, according to IEC 60364: Low voltage electrical installation, part 5-52: Selection and erection of electrical equipment - wiring Systems. Figure 5. Magnetic flux den sity Figure 6. V1 vs Ia Figure 7. Pout vs Ia Figure 8. η vs Ia -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0 10 20 30 40 50 60 70 80 90 100 B m g , T Tangential length of the air gap, mm No-load PF = 1 PF = 0.85 222 224 226 228 230 232 234 236 238 0 1 2 3 4 5 6 7 V 1 , (V ) Ia, (A) PF = 1 PF = 0.85 0 200 400 600 800 1000 1200 1400 0 1 2 3 4 5 6 7 P o u t, ( W ) Ia, (A) PF = 1 PF = 0.85 0 10 20 30 40 50 60 70 80 90 100 0 1 2 3 4 5 6 7 η (% ) Ia, (A) PF =1 PF = 0.85 K. Wirtayasa et al. / Journal of Mechatronics, Electrical Power, an d Vehicular Technology 10 (2019) 17–23 23 IV. Conclusion Load characteristic analysis of the double-side internal coreless stator AFPMG has been discussed in this paper. The applied load is resistive and resistive- inductive in series, which gives the power factor of 1 and 0.85 respectively. From the simulation, it is found that when loaded with resistive load, the coreless generator delivers higher armature current but this gives a consequent in higher voltage drop indicated by lower terminal voltage and higher voltage regulation. Nevertheless, with a better power factor, the output power and efficiency are higher. It is opposite to the generator that is loaded with impedance. According to the results, it can be concluded that the coreless generator performance is superior in the output power (1,241 W) and efficiency (91 %) with resistive load; on the other hand, the voltage regulation is better (5.86 %) with impedance load. From each parameter being compared, the difference in values is relatively small, so in principle, the generator provides good performance in both loads. Acknowledgement The authors would like to thank all the facilities provided by the Indonesian Institute of Sciences (LIPI), particularly to the Research Center for Electrical Power and Mechatronics during the process of making this manuscript. Declarations Author contribution K. Wirtayasa and P. Irasari contributed equally as the main contributor of this paper. All authors read and approved the final paper. Funding statement This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflict of interest The authors declare no conflict of interest. Additional information No additional information is available for this paper. References [1] S. S. Soe and Y. A. Oo, “Design of the Coreless Axial-Flux Double- Sided Permanent Magnet Synchronous Generator for Wind Power System,” Int. J. Sci. Eng. Technol. Res., vol. 3, no. 10, pp. 2047–2051, 2014. [2] S. Saint Soe and Y. A. N. A. Oo, “Design and Implementation of the Double-Sided Axial-Flux PMSG with Slotted Stator by Using Sizing Equation and FEA Software,” in Proceedings of Nineteenth TheIIER International Conference, 2015, no. April, pp. 63–69. [3] D. Ahmed and A. Ahmad, “An optimal design of coreless direct- drive axial flux permanent magnet generator for wind turbine,” J. Phys. Conf. Ser., vol. 439, 2013. [4] A. Daghigh, H. Javadi, and H. Torkaman, “Improved design of coreless axial flux permanent magnet synchronous generator with low active material cost,” in The 6th International Power Electronics Drive Systems an d Technologies Conference (PEDSTC2015), 2015, pp. 532–537. [5] S. S. Bagesh war, P. V Phan d, R. V Phan d, P. G. Student, an d U. G. Student, “Design & Analysis of Axial Flux Permanent Magnet Synchronous Generator,” Int. J. Res. Tren ds Innov., vol. 2, no. 7, pp. 123–129, 2017. [6] M. Chirca, S. Breban, C. A. Oprea, and M. M. Radulescu, “Analysis of Innovative Design Variations for Permanent-Magnet Generators in Micro-Wind Power Applications,” in 2014 International Conference on Electrical Machines (ICEM), 2014, pp. 385–389. [7] A. O. Otuoze, O. O. Ogidi, O. O. Mohammed, A. A. Jimoh, and B. S. Olaiya, “Sizin g of Wind Powered Axial Flux Permanent Magnet Alternator Using Analytical Approach,” Niger. J. Technol., vol. 35, no. 4, pp. 919–925, 2016. [8] M. R. Minaz and M. Çelebi, “Design and analysis of a new axial flux coreless PMSG with three rotors and double stators,” Results Ph ys., vol. 7, pp. 183–188, 2017. [9] M. R. Minaz and M. Çelebi, “Analysis and Design of an Axial-Flux Coreless Permanen t Magnet Synchronous Generator with Single Stators and Double Rotors,” Int. J. Energy Appl. Technol., vol. 4, no. 1, pp. 7–11, 2017. [10] D. Chun g and Y. You, “Design and Performance Analysis of Coreless Axial-Flux Permanent-Magnet Generator for Small Wind Turbines,” J. Magn., vol. 19, no. 3, pp. 273–281, 2014. [11] J. F. Gieras, R. Wang, and M. J. Kamper, “Axial Flux Permanent Magnet Brushless Machines”, 2nd ed. Springer, 2008. [12] S. Kim, J. Li, D. Choi, and Y. Cho, “Design and analysis of Axial Flux Permanent Magnet Generator for Direct-Driven Wind Turbines,” Int. J. Power Syst., vol. 2, pp. 1–6, 2017. [13] T. S. El-hasan, “Development of Automotive Permanent Magnet Alternator with Fully Controlled A C / DC Converter,” Energies, vol. 11, no. 274, pp. 1–28, 2018. Figure 9. VR vs Ia 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 V R , (% ) Ia, (A) PF = 1 PF = 0.85 http://ijsetr.com/uploads/432156IJSETR1290-350.pdf http://ijsetr.com/uploads/432156IJSETR1290-350.pdf http://ijsetr.com/uploads/432156IJSETR1290-350.pdf http://ijsetr.com/uploads/432156IJSETR1290-350.pdf http://www.worldresearchlibrary.org/up_proc/pdf/22-142815096163-69.pdf http://www.worldresearchlibrary.org/up_proc/pdf/22-142815096163-69.pdf http://www.worldresearchlibrary.org/up_proc/pdf/22-142815096163-69.pdf http://www.worldresearchlibrary.org/up_proc/pdf/22-142815096163-69.pdf http://www.worldresearchlibrary.org/up_proc/pdf/22-142815096163-69.pdf https://doi.org/10.1088/1742-6596/439/1/012039 https://doi.org/10.1088/1742-6596/439/1/012039 https://doi.org/10.1088/1742-6596/439/1/012039 https://doi.org/10.1109/pedstc.2015.7093331 https://doi.org/10.1109/pedstc.2015.7093331 https://doi.org/10.1109/pedstc.2015.7093331 https://doi.org/10.1109/pedstc.2015.7093331 https://doi.org/10.1109/pedstc.2015.7093331 https://pdfs.semanticscholar.org/5c83/add99d323b2563e019cbb8e681fd6d1b8673.pdf https://pdfs.semanticscholar.org/5c83/add99d323b2563e019cbb8e681fd6d1b8673.pdf https://pdfs.semanticscholar.org/5c83/add99d323b2563e019cbb8e681fd6d1b8673.pdf https://pdfs.semanticscholar.org/5c83/add99d323b2563e019cbb8e681fd6d1b8673.pdf https://doi.org/10.1109/icelmach.2014.6960209 https://doi.org/10.1109/icelmach.2014.6960209 https://doi.org/10.1109/icelmach.2014.6960209 https://doi.org/10.1109/icelmach.2014.6960209 https://doi.org/10.1109/icelmach.2014.6960209 https://doi.org/10.4314/njt.v35i4.29 https://doi.org/10.4314/njt.v35i4.29 https://doi.org/10.4314/njt.v35i4.29 https://doi.org/10.4314/njt.v35i4.29 https://doi.org/10.1016/j.rinp.2016.10.026 https://doi.org/10.1016/j.rinp.2016.10.026 https://doi.org/10.1016/j.rinp.2016.10.026 https://www.semanticscholar.org/paper/Analysis-and-Design-of-an-Axial-Flux-Coreless-with-Minaz-%C3%87elebi/d59219d7f391c0437220bbad53a78981e703a23c https://www.semanticscholar.org/paper/Analysis-and-Design-of-an-Axial-Flux-Coreless-with-Minaz-%C3%87elebi/d59219d7f391c0437220bbad53a78981e703a23c https://www.semanticscholar.org/paper/Analysis-and-Design-of-an-Axial-Flux-Coreless-with-Minaz-%C3%87elebi/d59219d7f391c0437220bbad53a78981e703a23c https://www.semanticscholar.org/paper/Analysis-and-Design-of-an-Axial-Flux-Coreless-with-Minaz-%C3%87elebi/d59219d7f391c0437220bbad53a78981e703a23c https://doi.org/10.4283/jmag.2014.19.3.273 https://doi.org/10.4283/jmag.2014.19.3.273 https://doi.org/10.4283/jmag.2014.19.3.273 https://link.springer.com/book/10.1007/1-4020-2720-6 https://link.springer.com/book/10.1007/1-4020-2720-6 https://www.iaras.org/iaras/journals/caijps/design-and-analysis-of-axial-flux-permanent-magnet-generator-for-direct-driven-wind-turbines https://www.iaras.org/iaras/journals/caijps/design-and-analysis-of-axial-flux-permanent-magnet-generator-for-direct-driven-wind-turbines https://www.iaras.org/iaras/journals/caijps/design-and-analysis-of-axial-flux-permanent-magnet-generator-for-direct-driven-wind-turbines https://doi.org/10.3390/en11020274 https://doi.org/10.3390/en11020274 https://doi.org/10.3390/en11020274