Design of a wind turbine generator for rural applications
2019; Institution of Engineering and Technology; Volume: 13; Issue: 3 Linguagem: Inglês
10.1049/iet-epa.2018.5734
ISSN1751-8679
AutoresCristian Ruschetti, Carlos Verucchi, Guillermo R. Bossio, Guillermo García, Matías Meira,
Tópico(s)Magnetic Bearings and Levitation Dynamics
ResumoIET Electric Power ApplicationsVolume 13, Issue 3 p. 379-384 Research ArticleFree Access Design of a wind turbine generator for rural applications Cristian Ruschetti, Cristian Ruschetti orcid.org/0000-0002-3183-7247 UNCPBA, INTELYMEC - CIFICEN (UNCPBA, CICPBA, CONICET), Del Valle 5737, Olavarría, ArgentinaSearch for more papers by this authorCarlos Verucchi, Corresponding Author Carlos Verucchi verucchi@fio.unicen.edu.ar UNCPBA, INTELYMEC - CIFICEN (UNCPBA, CICPBA, CONICET), Del Valle 5737, Olavarría, ArgentinaSearch for more papers by this authorGuillermo Bossio, Guillermo Bossio UNRC, GEA, ArgentinaSearch for more papers by this authorGuillermo García, Guillermo García UNRC, GEA, ArgentinaSearch for more papers by this authorMatias Meira, Matias Meira UNCPBA, INTELYMEC - CIFICEN (UNCPBA, CICPBA, CONICET), Del Valle 5737, Olavarría, ArgentinaSearch for more papers by this author Cristian Ruschetti, Cristian Ruschetti orcid.org/0000-0002-3183-7247 UNCPBA, INTELYMEC - CIFICEN (UNCPBA, CICPBA, CONICET), Del Valle 5737, Olavarría, ArgentinaSearch for more papers by this authorCarlos Verucchi, Corresponding Author Carlos Verucchi verucchi@fio.unicen.edu.ar UNCPBA, INTELYMEC - CIFICEN (UNCPBA, CICPBA, CONICET), Del Valle 5737, Olavarría, ArgentinaSearch for more papers by this authorGuillermo Bossio, Guillermo Bossio UNRC, GEA, ArgentinaSearch for more papers by this authorGuillermo García, Guillermo García UNRC, GEA, ArgentinaSearch for more papers by this authorMatias Meira, Matias Meira UNCPBA, INTELYMEC - CIFICEN (UNCPBA, CICPBA, CONICET), Del Valle 5737, Olavarría, ArgentinaSearch for more papers by this author First published: 27 February 2019 https://doi.org/10.1049/iet-epa.2018.5734Citations: 1AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract This paper presents a wind-generation system design for rural applications. In Southern areas of Argentina, there are right conditions for wind energy exploitation. Depending on the characteristics of these areas, the wind-generation systems can operate either isolated or connected to the grid. This work particularly emphasises the design of a generator to optimise the system performance and achieve its minimum volume. The analytical design is evaluated using finite element analysis and experimentally validated using a 35 kW prototype. The tests reveal that the design procedure is suitable. The initially proposed efficiency aim was exceeded, reaching a maximum value of 94.2%. The generator operating conditions make it particularly suited for the application for which it was designed. 1 Introduction Numerous studies have revealed that Patagonia, a region located at the southern end of Argentina, is a suitable area for the exploitation of wind resources [1]. In [2] are presented annual wind speed variations for this area. The average speed is 7.7 m/s with a 40% probability of occurrence. These values are above the minimum one required to guarantee the viability of power companies to produce electric energy. Although these areas are far from major consumption centres, it is possible to connect wind power generators to the grid or, as proposed in [3], to combine wind and other energy sources to reduce CO2 emissions. There is another possibility to exploit wind resources in the region through isolated small plants. This is the alternative presented in this work. That is, from the characteristics of the region (very low population density, large number of small rural settlements and low annual rainfall), it arises the necessity of using generation systems able to supply energy to both small isolated consumption areas and those grouped in wind farms, while supplying energy to the power grid. It can be concluded that the region under analysis presents suitable characteristics for the installation of low-power wind turbines able to: (a) Feed small isolated plants through diesel-wind hybrid systems of up to 100 kW. (b) Feed water pumping systems for irrigation. (c) Provide a system for the 'self-powered' market. That is, for individuals to supply their own consumption and supply the exceeding energy to the power grid. (d) Regulate voltage on weak lines, especially in rural areas. (e) Facilitate the interconnection of small- and medium-sized power systems, both direct current (DC) and alternating current (AC) systems, such as plants and hydrogen fuel cells or photovoltaic cells, and three-phase generators of primary energy sources, as for instance diesel, hydro, or gas, respectively. In [2], it is presented a comprehensive study on the possibilities to generate wind energy in Patagonia. This study proposes a methodology to find the most suitable areas for energy exploitation. In [4], it is presented the design of low-power wind generators for rural purposes and compares the different design possibilities depending on the characteristics of the turbine. In addition, it is important to notice that the systems used to generate electricity from wind power are generally composed of a three-blade horizontal axis turbine plus an electric generator. A wind turbine is designed to produce power over a range of wind speeds. The cut-in speed is around 3–4 m/s for most turbines, and it operates up to a rated speed of 12 to 15 m/s with a fixed step. If the rated wind speed is exceeded, the power has to be limited and one way to achieve this is by using the blade pitch control. In order to protect the mechanical system, the speed cut-out is around at 25 m/s [5]. The optimum operation speed values of wind turbines are below those of electric generators such as doubly fed induction machines. This requires the use of gear boxes as speed multipliers between the turbines and generators. These gear boxes increase the system costs, producing in addition a decrease in its overall performance. They also require more maintenance due to they are the major cause of failure in wind turbines, producing also unwanted noise [6, 7]. Since the nineties, designers have proposed some generators directly coupled to the wind turbines [8, 9], for which it is necessary to build low-speed generators with a high number of magnetic poles. An alternative to build low-speed generators is the use of permanent magnet electrical machines with rare earth magnets of high energy, such as NdFeB alloys (Neodymium -Iron-Boron). A reduction of volume and weight of the generators and its improvements in their performance are some of the advantages of using permanent magnet synchronous machines (PMSM). Most PMSMs use surface-mounted magnets in an inner rotor [6]. Chen et al., on the other hand, proposes the use of axial-flow geometries of permanent magnet (PM) assembly, which are appropriate for high number of poles [10]. Another alternative is proposed by Olano [11] and Choi et al. [12], which consists in mounted PMs on an outer rotor. This installation can be simplified by coupling the turbine blades directly to the outer rotor. Regarding the type of windings used to build the generator, there are several possible alternatives. A general classification of stator windings for PMSM distinguishes between distributed and concentrated windings. The concentrated windings can be subdivided into double-layered overlapped and single-layered non-overlapped. Concentrated windings presents advantages over distributed ones since they offer a reduction in copper volume and losses due to Joule effects, as well as improvements in performance and an increase in power density [13]. Finally, it is necessary to note that a phenomenon known as 'cogging torque' is common in PMSM. Such phenomenon is the product of the magnetic interaction between the rotor poles and the stator teeth. The number of equilibrium positions of the rotor is defined as the least common multiple between the number of slots and the number of poles. Though the average value of reluctance torque is zero, its presence can result in high starting torques and levels of vibration, for which it is recommended to reduce the cogging torque. Different solutions have been proposed that either modify the profile of PMs or fractionate them into longitudinal sections to reduce this effect [14]. This work aims to present the general characteristics of the proposed installation and specifically the permanent-magnet synchronous generator (PMSG) design. In the following sections of this paper, the general characteristics of the proposed installation are presented as well as the design of the generator. A verification using finite elements analysis (FEA) is carried out. Finally, experimental results from a build prototype are obtained. 2 General installation characteristics The power characteristic of the direct drive wind turbine developed is presented in Fig. 1. For this particular case, speed values are within the range 6 to 12 m/s without pitch control. Above 12 m/s, the blades are re-oriented in order to keep a constant speed with the turbine rotating at about 150 rpm. If speed values exceed 25 m/s, the blades orientate so that the turbine stops rotating and turns off. All these speed values are defined based on the wind characteristics in Patagonia areas. For this operation mode, the generator output voltages vary directly with the turbine rotation speed. The frequencies of the generated voltages are also variable depending on speed variations. To achieve a three-phase output system with constant voltage and frequency, an electronic converter is generally used. It rectifies the generator output voltage and then generates a useful AC frequency. In this application, Argentina grid-rated values are adopted (380 V and 50 Hz). Fig. 1Open in figure viewerPowerPoint Power curve of the wind turbine developed A scheme of the proposed installation is presented in Fig. 2. The energy produced by the generator is rectified and then converted into AC at the grid frequency. The DC bus allows energy storage in a battery bank. An additional option offers the possibility of including a conventional generator powered by a diesel engine. The maximum output power is fixed at 32 kW. Thus, the generator is designed for a nominal power of 35 kW. Fig. 2Open in figure viewerPowerPoint Installation Scheme As shown in [8], there is an optimal ratio between the tangential speed of the tip of a blade and the actual speed of the wind defined as 'tip speed ratio': (1) where ωr is the angular speed of the turbine, r is the radius of the blades and v, the wind speed. To reach optimal power coefficients in three-bladed turbines, it is necessary to achieve a value of λ = 7, then, with a wind speed of 12 m/s it is adopted a rotation speed of 150 rpm (as suggested in [8] for power of 35 kW). Then, r = 5.3 m. 3 Generator design 3.1 General dimensions The design of an electrical machine is an iterative process. This section presents the procedure to obtain the initial dimensions of the geometry and the windings. These dimensions are calculated using software that combines analytical calculation and heuristics procedures. Then, they are verified using FEA. Outer rotor topology was adopted due to the advantages it offers in the construction and assembly of the generator [6]. Regarding the stator windings, double-layer fractional pitch windings are used in order to reduce cogging torque and copper losses. To select the number of poles and generator slots, it becomes necessary to define the most suitable option with respect to the number of slots per pole and per phase (q) [15]: (2) where Q is the number of stator slots, m is the number of phases and p the number of poles of the generator. Values of q < 1 determine fractional pitch windings. This feature makes the coil heads shorter and allows reducing copper losses. For this particular design, it is proposed a value of q so as to obtain an electromotive force (EMF) with low harmonic distortion and therefore a waveform close to a sinusoidal one. A topology is proposed in which the fundamental winding factor k1 is close to 1. Finally, the possibility of limiting the cogging torque through the choice of q is discussed. Therefore, and based in [16], a value of q = 0.75 is adopted. The relationship between rotation speed and frequency, in a synchronous machine, is given by: (3) where f is the frequency of the induced voltages. In order to obtain a frequency near 50 Hz, it is chosen p = 36. Then, according to (2), Q = 81. According to [15], the dimensions of the machine air gap are directly dependant on, (4) where Snom is the nominal apparent power of the generator, η represents the efficiency, factor k1 is the stator winding factor, fa is the stacked plates, Dos is the stator outer diameter, ls, the length of the plate stacking, Bg1 is the amplitude of the fundamental component of flux density in the air gap and Ks is the linear density of current on the air-gap perimeter in A/m. In [15], there are suggested optimum values for the relationship between the stator outer diameter and the air gap length. For the proposed number of poles, the ratio is 2.5. Thus, from (4), it can be obtained the following dimensions: Dos = 683 mm and ls = 149 mm. Then, it is possible to obtain the slot pitch from the stator outer diameter as, (5) In [17], it is proposed dividing the slot pitch into equal parts to obtain the tooth and slot widths (w1 and ts in Fig. 3). Fig. 3Open in figure viewerPowerPoint Generator topology and general dimensions The PMs are arranged on the rotor inner surface, as illustrated in Fig. 3. If the PMs occupy an arc of 2α electrical degrees, then the maximum induction in the air gap may be calculated according to [18], as: (6) where Bg0 is the average induction in the air gap. The maximum value of Bg1 is 1.1 T, determined depending on the silicon saturation of the steel plates. The angle α is chosen equal to 60 electrical degrees, as recommended in [18], who demonstrates that around this value, there is an optimum point in term of cost. The mean value of flux density in the air gap (Bg0) is 0.998 T. Given Bg0, and choosing the tooth width ts equal to the slot width w1, then, the maximum tooth density becomes 1.995 T. 3.2 Permanent magnets design In order to optimise the machine performance through its design, it is proposed the use of NdFeB PMs, whose residual flux density is around 1.2 T. Its energy product is 300 kJ/m3 and can reach a safe operation temperature up to 150°C [19]. Based on Ampere's law, on a basic magnetic circuit, without magnetomotive force (MMF) and where the PM establishes the magnetic flux (7) where g is the air gap thickness, Hm is the PM magnetic field, lm is the PM height y Hg0 is the air gap magnetic field. Furthermore, if the cross-sections of the permanent magnets and air gap are equal to (8) where Bm is the PM induction. Taking into account the Bm–Hm linear characteristic of the NdFeB, in the working point of the PM [18] and knowing that the relative permeability of the material is ∼1 (9) where Br is the PM residual flux density (1.2 T for NdFeB). Getting Hm from (9), replacing it in (7) and considering the air gap permeance irregularity through the Carter coefficient,(Cc) the relation between the intended Bg0 and the PM residual induction is given by, (10) where lge is the gap effective length, given by, (11) Given that PM permeability in the direction of its magnetisation is practically equal to that of air, its height is added to that of the gap, as shown in (11). Moreover, considering that the air gap of the machine is modified by the stator slots, its length must be corrected using the Carter coefficient, whose value is chosen based in [18]. Thus, taking a g = 1.5 mm, Cc = 1.05 and Br = 1.2 T, yields lm = 10.3 mm. 3.3 Core design The core design consists of calculating the height of the circuit components for which the magnetic flux of stator and rotor closes. This is commonly called 'yoke' and requires knowing the flux per pole of the machine [18]: (12) The thickness of the stator core is dimensioned so that the flux density Bcs does not exceed 1 T. This density value is calculated as, (13) Then, the yoke height dcs must be 21.29 mm. Slots are sized considering its width (w1) and the tooth thickness (ts) equal. The depth of slots ds, as a first approximation, is chosen equal to five times the width w1. 3.4 Winding design It is adopted a double-layered overlapped winding. Since the winding considers a fractional value q, the coils of each phase should be located according to the diagram of Fig. 4. The beginnings and endings of coils for each phase are represented by A, B, and C and by −A, −B and −C, respectively. The winding step (y) is equal to 2. Fig. 4 only shows nine slots because this pattern repeats for the remaining stator slots. Fig. 4Open in figure viewerPowerPoint Scheme of the winding, 81slots, 36 poles To calculate the number of turns per coil and the wire diameter, it should be noted that the induced EMF at each phase, for a double-layered overlapped winding with all the coils in series, is given, according to [5], by: (14) where, Nb is the number of turns per coil. Given that the voltage drops at each phase cannot be known, and considering a desirable voltage at the terminals of 220 V per phase at nominal rate, it is initially assumed a no load voltage of 250 V. Therefore, the turns per coil result Nb = 8. Finally, the maximum wire diameter is calculated assuming a slot usage factor (far = 0.4). (15) 3.5 Design optimisation With all the preliminary dimensions so far obtained, it begins an iteration process with the purpose of adjusting these dimensions in order to reduce the generator volume while maintaining its efficiency above 92%. This efficiency value is determined based on typical values for this type of machines [15]. The optimisation process is carried out using the methodology proposed in [20]. The variables adjusted during the optimisation process are the rotor diameter, some slots dimensions, and the PMs length and height. The optimisation process was carried out maximising the efficiency and looking for keep bounded the structure weight. For this, the slot filling percentage is increased to reduce electrical losses. Moreover, avoiding magnetic induction concentration areas, the magnetic losses are limited. The dimensions adopted are shown in Table 1. Table 1. General dimensions of the designed machine Parameter Analytical res. RMxprt res Dos 682.85 mm 700 mm ls 148.97 mm 150 mm Ks 30 kA/m 28.353 kA/m λs 26.48 mm 27.15 mm g 1.5 mm 1.6 mm lm 10.30 mm 12 mm dm 39.73 mm 40 mm lpm 49.66 mm 50 mm r1 5 mm 6.33 mm r2 5 mm 3.62 mm dos 3 mm 3 mm wos 4 mm 5 mm ts 12.93 mm 13.8 mm w1 12.93 mm 12.66 mm ds 64.66 mm 70 mm w2 7.92 mm 7.24 mm Dis 484.95 mm 480 mm Dpm 685.85 mm 703.2 mm Dir 706.45 mm 727.2 mm Dor 749.03 mm 766 mm Nb 8 8 dcu 2.3 mm 2.3 mm Finally, the stator resistance and inductance can be calculated as well as voltage drops, losses, efficiency, cogging torque, and PM demagnetisation. The verification of all these values can be seen in [21]. 4 FEA verification For the FEA, bi-dimensional software is used since radial-flux laminated-stator machines allow obtaining satisfactory results avoiding high computational costs. As a first step, a magneto-static analysis is carried out with the objective of verifying the border conditions and the magnetisation direction of PMs. This software automatically generates the global equation system, which can be solved using either the Incomplete Cholesky Conjugate Gradient (ICCG) method or the direct elimination method for linear equations. Fig. 5 presents values for magnetic induction for a sector of the generator. These values are obtained through the FEA and match those of the analytical design with minor differences, no more than 3%. Fig. 6 presents the distribution of magnetic flux without load. Fig. 5Open in figure viewerPowerPoint Magnetic Induction obtained through FEA Fig. 6Open in figure viewerPowerPoint Magnetic field distribution on the generator Likewise, losses and machine performance are also verified, as well as internal voltage drops and cogging torque, showing that the obtained results practically match those estimated in the analytical design [21]. 5 Experimental verification A prototype was built to experimentally validate the proposed design. It can be observed in Fig. 7 the generator mounted on a testing bench that consists of a speed reducer and an induction motor used to drive the generator. The test bench also includes resistive loads. With it is application and using a load cell it is possible to measure the induced torque of the generator. Fig. 7Open in figure viewerPowerPoint Generator mounted on test bench Different tests were carried out on the generator to validate the proposed design. No-load tests show values for voltage drops of ∼440 V. These values show differences <1% compared to those predicted using the FEA. The voltage harmonics are about 1% the fundamental component for the 5th harmonic and 0.2% the fundamental component for the 7th harmonic. This determines a THD = 1.2%. Fig. 8 presents voltage regulation at the generator terminals for variations of load between no-load condition and 120% of nominal load. The voltage regulation for nominal load with unity power factor is ∼5%. During the design process, no restrictions are imposed on internal voltage drops, focusing on the machine performance. Given that the generator output voltages are rectified, voltage drops are acceptable. Fig. 8Open in figure viewerPowerPoint Voltage at the terminals vs. output power Results for the generator under load conditions are shown in Table 2. Stepwise loads were applied up to the nominal value. The generator efficiency, at rated speed, is 94%, higher than the estimated one. This value is comparable to the results for a 30 kW rated power and 48 poles PMSG presented in [22]. The generator efficiency for different loads with unity power factor can be observed in Fig. 9. The mean absolute percentage error (MAPE) between the experimental and the FEA results is 0.45%. This is a very acceptable value taking into account that in FEA model, the magnetic losses were carried out through a function. The function coefficients were obtained for the whole induction working range of the core material. Furthermore, the friction losses were estimated from typical values. Table 2. Experimental results Variable LOAD 6 [kW] 12 [kW] 18 [kW] 24 [kW] 30 [kW] input torque [Nm] 607.70 1106.76 1612.36 2047.72 2447.95 input power [kW] 9.68 17.62 25.66 32.62 38.96 output power [kW] 8.30 16.30 23.95 30.71 36.70 ɳ (%) 85.7 92.5 93.3 94.2 94.2 cos(φ)Load 1 1 1 1 1 phase current [A] 10.88 21.69 31.90 41.74 51.10 unbalance I [%] 2.62 2.23 1.02 0.96 1.01 THD I [%] 1.86 1.73 1.63 1.61 1.53 line voltage [V] 453.75 447.74 439.82 431.67 421.63 unbalance V [%] 0.72 0.73 0.84 0.86 0.87 THD V [%] 1.36 1.48 1.56 1.53 1.45 f [Hz] 45.64 45.63 45.60 45.64 45.60 Fig. 9Open in figure viewerPowerPoint Efficiency vs. output power The increase of temperature for nominal load is 80°C. Fig. 10 presents a thermal infrared image that corresponds to one of the tests. It is important to notice that tests were carried out under disadvantageous cooling conditions, in complete absence of ventilation. Under normal operation conditions, and taking advantage of wind refrigeration, the temperature jumping results always lower than that estimated ones. Fig. 10Open in figure viewerPowerPoint Temperature reached by the generator components (a) Windings, (b) Permanent magnets Fig. 11 shows the temperature evolution inside a slot in the upper part of the generator for different load values. It is important to verify the temperature inside the magnets since if it exceeds certain values, then residual magnetism losses will occur and they will be irreversible. In [23, 24], residual magnetism losses at any generator pole is deeply analysed and techniques of easy implementation for faults detection due to loss of magnetism are proposed. Fig. 12 presents the induced EMF variation due to temperature. The experimental result is approximate with a linear least squares fitting, with an acceptably low level of root mean square error (RMSE) of 5.26 × 10−4 and a coefficient of determination R2 = 0.9904. Therefore, the linear fitting correctly represents the analysed characteristic. Knowing the temperature effect over the EMF allows predicting its reduction under operation conditions. The linear approximation gives a 0.045% FEM reduction rate, while the Br reduction rate for the NdFeB PM used in this prototype is three times higher. The difference between the reduction rates is due to the magnetic core saturation. This characteristic results advantageous for voltage regulation. Fig. 11Open in figure viewerPowerPoint Evolution of temperature inside the slots Fig. 12Open in figure viewerPowerPoint Reduction of the induced EMF due to temperature effects 6 Conclusions A proposal for the exploitation of wind energy in areas with particular characteristics was presented. This proposal consists in designing and building installations that aim to meet small and medium loads demands, provide energy for artificial irrigation, supply the exceeding energy to the power grid, and compensate voltage drops in lines. The electric generator was designed to work directly coupled to the turbine. It consists of a PM outer rotor and fractional pitch windings. A geometry optimisation was carried out by analysing the analytical design using FEA, with the objective of maximising the machine efficiency and reducing its volume. The experimental tests allow concluding that the design procedure is suitable. Understanding that the procedure combines analytical and FEA design, it was reached a 94.2% generator efficiency, with a 0.45% MAPE respect off the numerical model. This high efficiency and the generator operating conditions make it particularly suited for the application for which it was designed. This proposal is an alternative to conventional methods to produce and take advantage of wind energy in rural areas of low population. 7 Acknowledgments This research was supported by the ANPCYT, CONICET, Universidad Nacional del Centro de la Provincia de Buenos Aires and Universidad Nacional de Río Cuarto (Argentina). 8 References 1Lássig, J., Cogliati, M., Bastanski, M., et al: 'Wind characteristics in neuqen, north Patagonia, Argentina', Wind Eng. Ind. Aerodyn., 1999, 79, pp. 183– 199 2Oliva, R.: 'Simulation and measurement for effective isolated wind hybrid system development in south Patagonia', Energy. Sustain. Dev., 2008, 12, (2), pp. 17– 26 3Spazzafumo, G.: 'South Patagonia: wind/hydrogen/coal system with reduced CO2 emissions', Hydrog. Energy, 2013, 38, pp. 7599– 7604 4Al-Bahadly, I.: 'Building a wind turbine for rural home', Energy. Sustain. Dev., 2009, 13, (3), pp. 159– 165 5Gönen, T.: ' Electrical machines with MATLAB' ( CRC Press, United States, 2012, 2nd edn.) 6Ki-Chan, K., Seung-Bin, L., Ki-Bong, J., et al: ' Analysis on the direct-driven high power permanent magnet generator for wind turbine'. Proc. Eighth Int. Conf. on Electrical Machines and Systems, Nanjing, China, 2006, pp. 243– 247 7Fengxiang, W., Jianlong, B., Qingming, H., et al: ' Design features of low speed permanent magnet generator direct driven by wind turbine'. Proc. Eighth Int. Conf. on Electrical Machines and Systems, Nanjing, China, 2006, pp. 1017– 1020 8Chen, J., Nayar, C., Xu, L.: ' design and finite-element analysis of an outer-rotor permanent-magnet generator for directly coupled wind turbines', IEEE Trans. Magn., 2000, 36, (5), pp. 3802– 3809 9Khan, M., Pillay, P., Malengret, M.: ' Impact of direct-drive WEC systems on the design of a small PM wind generator'. Proc. IEEE Power Tech. Conf., Bologna, Italia, 2003, pp. 7– 13 10Chen, Y., Pillay, P.: ' Axial-flux PM wind generator with a soft magnetic composite core'. Proc. Industry Applications Conf., Hong Kong, China, 2005, pp. 231– 237 11Olano, A., Moreno, V., Molina, J., et al: ' Design and construction of an outer-rotor PM synchronous generator for small wind turbines; comparing real results with those of FE model'. Proc. Electrical Machines, ICEM, Sydney, Australia, 2008, pp. 1– 6 12Choi, J., Jang, S.-M, Song, B.-M.: ' Design of a direct-coupled radial-flux permanent magnet generator for wind turbines'. IEEE Power and Energy Society General Meeting, Minnesota, USA, 2010, pp. 1– 6 13El-Refaie, A.: 'Fractional-slot concentrated-windings synchronous permanent magnet machines: opportunities and challenges', IEEE Trans. Ind. Electron., 2010, 57, (1), pp. 107– 121 14Islam, R., Husain, I., Fardoun, A., et al: 'Permanent-magnet synchronous motor magnet designs with skewing for torque ripple and cogging torque reduction', IEEE Trans. Ind. Appl., 2009, 45, (1), pp. 152– 160 15Lipo, T.: ' Introduction to AC machine design' ( Wiley IEEE Press, United States, 2017) 16Hanselman, D.: ' Brushless permanent magnet motor design' ( Electrical and Computer Engineering University of Maine, USA, 2006) 17Pyrhönen, J., Jokinen, T., Hrabovcová, V.: ' Design of rotating electrical machines' ( John Wiley & Sons Ltd., Great Britain, 2009) 18Slemon, G.: 'On the design of high-performance surface-mounted PM motors', IEEE Trans. Ind. Appl., 1994, 30, (1), pp. 134– 140 19Sebastian, T.: 'Temperature effects on torque production and efficiency of PM motors using NdFeB magnets', IEEE Proc. Ind. Appl., 1995, 31, (2), pp. 353– 357 20He, Q., Wang, Q.: ' Optimal design of low-speed permanent magnet generator for wind turbine application'. Proc. Power and Energy Engineering Conf., Shanghai, China, 2012, pp. 1– 3 21Ruschetti, C.: ' Diseño de generadores sincrónicos de imanes permanentes de velocidad variable para turbinas eólicas', Phd. Thesis, Universidad Nacional de Río Cuarto Press, 2012 22Bazzo, T., Kolzer, J.F., Carlson, R., et al: ' Optimum design of a high-efficiency direct-drive PMSG'. 2015 IEEE Energy Conversion Congress and Exposition (ECCE), Montreal, QC, 2015, pp. 1856– 1863 23Ruschetti, C., Bossio, G., De Angelo, C., et al: ' Effects of partial rotor demagnetization on permanent magnet synchronous machines'. Proc. IEEE Int. Conf. on Industrial Technology, Viña del Mar, Chile, 2010, pp. 1233– 1238 24Ruschetti, R., Verucchi, C., Bossio, G., et al: 'Rotor demagnetization effects on permanent magnet synchronous machines', Energy Convers. Manage., 2013, 74, pp. 1– 8 Citing Literature Volume13, Issue3March 2019Pages 379-384 FiguresReferencesRelatedInformation
Referência(s)