Evaluation of a magnetic gear for air‐cooled condenser applications
2018; Institution of Engineering and Technology; Volume: 12; Issue: 5 Linguagem: Inglês
10.1049/iet-epa.2017.0714
ISSN1751-8679
AutoresAlexander Matthee, Rong‐Jie Wang, Charles Johannes Agenbach, Daniel N.J., Maarten J. Kamper,
Tópico(s)Magnetic Properties and Applications
ResumoIET Electric Power ApplicationsVolume 12, Issue 5 p. 677-683 Research ArticleFree Access Evaluation of a magnetic gear for air-cooled condenser applications Alexander Matthee, Alexander Matthee Department of Electrical and Electronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorRong-Jie Wang, Corresponding Author Rong-Jie Wang rwang@sun.ac.za Department of Electrical and Electronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorCharles J. Agenbach, Charles J. Agenbach Department of Mechanical and Mechatronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorDaniel N.J. Els, Daniel N.J. Els Department of Mechanical and Mechatronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorMaarten J. Kamper, Maarten J. Kamper Department of Electrical and Electronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this author Alexander Matthee, Alexander Matthee Department of Electrical and Electronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorRong-Jie Wang, Corresponding Author Rong-Jie Wang rwang@sun.ac.za Department of Electrical and Electronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorCharles J. Agenbach, Charles J. Agenbach Department of Mechanical and Mechatronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorDaniel N.J. Els, Daniel N.J. Els Department of Mechanical and Mechatronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this authorMaarten J. Kamper, Maarten J. Kamper Department of Electrical and Electronic Engineering, Stellenbosch University, Stellenbosch, 7600 South AfricaSearch for more papers by this author First published: 09 April 2018 https://doi.org/10.1049/iet-epa.2017.0714Citations: 6AboutSectionsPDF 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 In high-power industrial applications such as air-cooled condenser fans, excessive mechanical gear failures highlight a need to investigate possible alternatives. In this study, the concentric magnetic gear (CMG) is evaluated as a potential replacement for traditional mechanical gear. For a more objective comparison between the CMG and mechanical gear, a CMG is optimally designed in accordance with the performance specifications of an existing mechanical gear. The design is further refined for performance improvements taking into account mechanical strength requirements. Based on the design a CMG prototype is manufactured and experimentally evaluated against its mechanical counterpart under the same operating conditions. The performance of both gears achieved efficiency in the mid 90% range. The mechanical gear slightly outperformed the CMG under rated operating conditions. The large unbalanced magnetic forces within the CMG are likely responsible for the larger than expected no-load losses. With the advantage of overload protection, reduced maintenance requirements, the CMG could be a valid replacement for the mechanical gear. 1 Introduction In power industry, the majority of recent capacity addition in power generation is renewable energy based. Among various new generation power plant technologies, the concentrated solar power (CSP) plants are gaining momentum due to their potential high capacity factor and comparatively low power generation costs. Since these CSP plants are often located in water scarce and desert-like regions, the use of dry cooling for power plant operation is essential. A dry-cooled power plant usually uses a large number of air-cooled condensers (ACC). Fig. 1 shows the configuration of an A-shaped ACC, which consists of a steam-pipe and radiator assembly, a mechanically geared induction motor and a large axial fan [1]. Owing to the high starting torque [2], fan vibration [3], the impact of load fluctuations and cross air flows [4] that the mechanical fan drive system experiences, high failure rate of mechanical gears has been reported [5, 6]. In recent years, concentric magnetic gears (CMGs) have received increased attention from research institutions and industry [7-9]. Inherent qualities of magnetic gears such as no frictional contact between input and output shafts, potential high efficiencies and overload protection show promise as a viable replacement of mechanical gears in certain applications such as ACC fans. Despite that extensive research has been conducted on CMGs, there is little published work on the comparative study between the mechanical and magnetic gears [10]. In this paper, the magnetic gear is evaluated as an alternative to mechanical gear in an attempt to provide a reliable and low maintenance solution for dry-cooling CSP plant fan systems. Fig. 1Open in figure viewerPowerPoint Configuration of an A-shaped ACC system [1] 2 Design specifications The standard fan drive configuration in large ACC systems consists of an induction motor coupled with a helical type mechanical gearbox. Comparing with mechanical gears, there are only a few effective magnetic gear topologies that can compete with their mechanical counterpart, of which the CMG is mechanically least challenging to realise. Although the operating principle of CMG may be closer to that of planetary mechanical gear [10, 11], the main objective of this work is to evaluate the magnetic gear as a possible replacement technology for these ACC systems. As a design base, an existing single-stage helical mechanical gear is chosen, which has a gear ratio of 3.78:1, a rated torque capability of 132 N m and a service factor of 1.75. The specifications of the mechanical gear are listed in Table 1, where the mass and volumetric torque densities are calculated based on the rated torque. The CMG will be designed with equivalent ratings and specifications. Table 1. Specifications of the mechanical and magnetic gears Description Mechanical gear Magnetic gear rated torque, N m 132 132 gear ratio 3.78:1 3.75:1 service factor 1.75 1.75 maximum torque, N m 231 254 input speed, rpm 600 600 output speed, rpm 158 160 space envelope (L × W × H), mm 324 × 272 × 324 — volume (excl. shafts), m3 — total mass, kg 52 — mass torque density, N m/kg 2.54 — volume torque density, kN m/m3 10.74 — 3 Design procedure The basic operation principle of a CMG is that the ferromagnetic pole pieces (also called flux modulator) interposed between the inner and outer permanent magnet (PM) carriers modulate the magnetic field such that each PM carrier 'sees' a working space harmonic corresponding to its own number of poles. The number of ferromagnetic pole pieces, , and the number of PM pole pairs on the inner and outer PM carriers, and , must satisfy a specific relationship in order to work with these asynchronous magnetic harmonics. The angular velocity of the inner PM carrier, the outer PM carrier and the flux modulator are represented by , and , respectively. The relationship between the three components is governed by (1) where . Any two of the components can be selected as the input and output shafts. In this design, the outer PM carrier is selected to be stationary, therefore, . The gear ratio is represented by (2) To realise a gear ratio close to that of the mechanical gear, a number of pole-pairs options are available as shown in Table 2. For a given speed, the larger the pole-pairs number is, the higher the frequency-related core loss becomes. In addition, the cogging/ripple torques of a CMG should also be small as they contribute to the noise and vibration during the gear operation. A cogging torque factor [12], expressed in (3), can be used to estimate the severity of cogging/ripple torque (3) where p represents the number of pole-pairs of outer or inner PM carrier. A higher least common multiple (LCM) of and leads to a smaller cogging torque factor. Given the above considerations, the design option 4 is chosen, which results in a gear ratio of 3.75:1. The specifications of the CMG are also given in Table 1. Table 2. Design options for the MG design Option 1 2 5 7 3.5 1 2 3 8 11 3.67 1 3 4 10 14 3.5 2 4 4 11 15 3.75 1 5 5 13 18 3.6 2 6 5 14 19 3.8 1 The bold values indicate the chosen design. 3.1 Design optimisation The rated torque of the CMG is specified as 132 N m with a service factor of 1.75. Allowing for a 10% operational margin, the maximum torque (stall torque) of the CMG is thus 254 N m. For simplicity, the design calculation of the CMG is usually conducted by using finite-element (FE) modelling at the maximum torque angle position (), in which the angular relationship between the modulator and the inner and outer PM carriers as shown in Fig. 2 satisfy [13] (4) Fig. 2Open in figure viewerPowerPoint Offset angles and their definitions in a CMG To account for the end-effects of CMGs, three-dimensional (3D) FE modelling is preferred. However, 3D FE simulation is computationally expensive, especially in an iterative design optimisation procedure. Alternatively, 2D FE simulation can be employed to find an optimum design, which is calibrated using an end-effect ratio . The factor is defined as the stall torque ratio between 3D and 2D FE results [14] (5) Depending on the quality of designs, the end-effect ratio of some historic CMG prototypes ranges from 0.6 to 0.8 [8, 12, 15-17]. For this design, a conservative is assumed, which takes into account the effects of 3D end field and further fine-tuning measures. The optimisation objective function is formulated to maximise the stall torque of the CMG per volume, which is subjected to the torque constraint as shown in the following equation: (6) The design variables are illustrated in Fig. 3. To comply with the same space envelope as that of the mechanical gear, the maximum outer diameter and stack length of the CMG are constrained as 272 and 324 mm, respectively. Considering the manufacturability, large radial forces and associated deflection on the flux modulator an air-gap length of 1 mm is selected. Fig. 3Open in figure viewerPowerPoint Design variables of the CMG The overall design optimisation process is described by the flow diagram shown in Fig. 4. VisualDOC software suite of VR&D Inc.is used as the optimisation toolbox that communicates with a in-house 2D FEMelectromagnetic package (called SEMFEM) for performance characterisation. Theoptimisation algorithm applied in this case is the modified method of feasibledirection (MMFD). Since MMFD is a gradient-based method, the optimisation wasrepeated with different starting values to avoid possible convergence to a localmaximum. The optimum design and the results are summarised in Table 3. Table 3. Optimised design parameters of the CMG Parameter(variable) Initial value Final value gear outer radius, mm 59.5 81.5 , mm 5.0 6.0 , mm 5.0 20.0 , mm 5.0 5.5 , mm 5.0 6.0 ippi, ratio ofsegment pitch 0.5 0.57 ippo, ratio ofsegment pitch 0.5 0.46 , ratio of polepitch 0.85 0.90 , ratio of polepitch 0.85 0.90 modulator (LS) thickness, mm 5.0 9.0 modulator bridge thickness, mm 0.5 0.65 LS peak torque (2D results),N m — 337 HS peak torque (2D results),N m — 89.87 Parameter(pre-defined) Value HS number of pole pairs 4 LS number of steel segments 15 PM carrier number of pole pairs 11 stack length, mm 100 magnet grade N48H outer air-gap width, mm 1.0 inner air-gap width, mm 1.0 gear ratio 3.75:1 Fig. 4Open in figure viewerPowerPoint Flow chart of the optimisation process 3.2 Mechanical design aspects To realise a reliable magnetic gear design, it is essential to investigate the mechanical strength and durability of the components, which entails Identify components that will endure the highest stress. Calculate position and magnitude of the maximum stresses. Calculate safety factors against failure. To mitigate the possible deflection caused by large radial forces on the flux modulator [18], the flux modulator was manufactured using inner-bridged laminations, which were reinforced by epoxy resin encapsulation as proposed in [19]. In addition, relatively large air-gaps were assumed in the design. According to FE simulations, the maximum torque on the high-speed (HS) and low-speed (LS) shafts was found to be 88 and 330 N m, respectively. As shown in Fig. 5, the part of the LS shaft where the highest stress occurs is the key-way area. Also seen from Fig. 5, the load path demonstrates the torque being transferred from the bolts to the LS shaft. Thus, the bolts were also analysed to ensure they will be able to endure the load. Fig. 5Open in figure viewerPowerPoint LS shaft design (a) Sectioned view showing load path and components, (b) MSC Patran FE strength analysis of the shaft support: (left) tensor stress results, (right) boundary conditions By applying the analysis method described in [20], first the shaft key-way will be analysed, followed by the bolts, and finally the shaft supports. All calculations are done with the assumption of pure shear stress (meaning only torque and no external forces applied). The safety factor , at the key-way, was obtained by using the modified-Goodman equation, given as (7) where is the ultimate tensile stress, is the endurance limit, is the amplitude stress and is the median stress. The safety factor for the LS shaft was found to be 1.36. In a similar manner, the safety factor of the HS shaft was calculated as 3.38. The shaft safety factor is considered to be acceptable as the torque values obtained from 2D FE simulations are generally overestimated. Also the modified-Goodman equation (7) is considered as a conservative method of calculating the fatigue safety factor compared to other methods. Next the bolt shear stresses are considered. For the bolt stress calculations a more realistic maximum torque was used to avoid excessive bolt sizes. As seen from Fig. 5a, the outer bolts compress and connect the shaft support to the modulator pole pieces, while the inner bolts connect the shaft support to the LS shaft. The inner bolts endure a higher stress than the outer ones, due to a reduced number of bolts at a smaller moment arm. Table 4 summarises the relevant dimensions and factors. The safety factors for the inner and outer bolts were calculated as 1.89 and 5.1, respectively. Table 4. Stress analysis on bolts Factor Value Description 8 number of bolts 45 mm radius of bolts 3.14 mm minor diameter of bolt 310 MPa yield stress (bolt steel grade 4.8) Finally, the stresses developed in the LS shaft support are considered. An FE tensor stress analysis was done on the LS shaft support. The boundary conditions can be seen in the right side of Fig. 5b, and the tensor stress plot on the left side. Acetal co-polymer was initially considered, which achieved a safety factor of 1.79. However, due to structural stiffness, a stronger material PEEK (polyetheretherketone) [21] was chosen for the final design. 4 Design and efficiency improvement measures 4.1 Loss calculation To determine the electromagnetic losses of the CMG, transient 2D FE motion solver of MagNet 7 was utilised. This included simulations of both hysteresis and eddy losses of the CMG at steady state. Since the modulator (as LS rotor) rotates 3.75 times slower than the HS rotor, two motion components with respective rotation speeds are defined in the 2D FE model. A simulation time window of 20 ms with time steps of 0.1 ms was used. The losses caused by eddy currents and hysteresis in the HS yoke, modulator, PM carrier and supporting bolts are given in Table 5. Table 5. Losses in the yokes and supporting bolts (at rated speed: 160 rpm) Component Eddy current losses, W Hysteresis losses, W HS yoke 3.48 0.01 LS (modulator) 0.556 4.8 PM carrier yoke 0.183 2.14 bolts (modulator) 0.037 — To reduce magnet losses, PMs on both the HS rotor and outer PM carrier were segmented. Fig. 6 illustrates the difference between un-segmented (Fig. 6a) and segmented (Fig. 6b) PMs. The effects of magnet segmentation and modulator thin bridges to the PM losses were also evaluated using FE analysis. It can be observed in Table 6 that (i) the introduction of modulator thin bridges reduces magnet losses by almost 12.6% at a cost of small reduction of stall torque (6.6%), (ii) magnet segmentation reduces the magnet losses and the stall torque of the CMG by 50 and 15%, respectively. Table 6. Losses for segmented versus un-segmented PMs Configuration PM losses, W , N m un-segmented PM (no bridges) 30.9 360.8 un-segmented PM (bridges) 27.02 337 segmented PM with bridges 13.44 294 Fig. 6Open in figure viewerPowerPoint PMs in (a) Original CMG design, (b) Final CMG design with segmented magnets (showing also the supporting bolts in the flux modulator) 4.2 Demagnetisation check NdFeB magnets are sensitive to high temperatures and the risk of demagnetisation increases at high working temperature [22]. In addition, the CMG's performance also decreases with an increase in PM temperature. Reducing the losses in the magnets is especially vital as this may affect the performance and lifetime of the machine. Considering that the losses in the gear are not negligible, the performance of the gear should be checked at higher temperatures. The demagnetisation analysis of the CMG was conducted at and , respectively. The simulation predicts no demagnetisation in the gear at as shown in Fig. 7, even when the gear is at its maximum load angle and the probability of demagnetisation is at the maximum. At higher temperatures the gear experiences some demagnetisation at the trailing edge of the outer carrier magnets [23, 24]. Fig. 7Open in figure viewerPowerPoint Demagnetisation plot of the CMG at (a) HS PMs, (b) Outer PMs With ohmic losses being more concentrated in the corners of the PMs, simply removing the corner sections reduces losses even further with a marginal effect on output torque. With this implemented the torque output reduced to ∼281 N m and PM ohmic losses reduced to 8 W. The total ohmic and hysteresis losses are shown in Table 7. Note this excludes mechanical losses. Table 7. Calculated core loss components in the MG design with shaped PMs Loss in component Loss, W PMs (ohmic) 8.0 laminated yokes (eddy current) 0.354 laminated yokes (hysteresis) 12.41 total 20.76 4.3 3D FEM analysis Efficiency improvement measures implemented during the design refinement phase may negatively affect stall torque of the CMG. A 3D FEA verification was performed giving a peak torque output of 254 N m. Due to the relatively long stack length of the design, a better actual end-effect ratio is realised, i.e. . 5 Manufacturing of the prototype The manufacturing process of a CMG is described in this section. The modulator (LS rotor) was assembled by compressing the laminated steel segments in an aluminium mould with a pair of PEEK end plates. To strengthen the structure, epoxy resin was applied in the space and cavities between the pole segments of the modulator inside a vacuum chamber. The LS shaft was then added to the component to complete the assembly shown in Fig. 8a. The completed HS rotor is shown in Fig. 8b, where the PMs are secured with special epoxy glue. Fig. 8c shows the partially completed assembly of the CMG. Fig. 8Open in figure viewerPowerPoint Images showing the assembled CMG and its components with (a) LS rotor, (b) HS rotor, (c) Final CMG assembly 6 Experimental investigation In this section, the performance characteristics of both the CMG prototype and an equivalent mechanical gear are experimentally evaluated and compared. 6.1 Experimental set-up The test set-up is shown in Fig. 9, which consists of a geared induction motor with a variable-speed drive (VSD) as the prime mover, two Lorenz torque sensors for input/output speed and torque measurements, and a second geared induction motor with VSD drive running in regenerative mode as the load. The gear under test is connected between the two torque sensors. Fig. 9Open in figure viewerPowerPoint Test bench set-up for (a) Mechanical gear, (b) Magnetic gear 6.2 No-load tests The no-load losses of both mechanical and magnetic gears were measured in the first test to benchmark the base losses. The no-load loss versus rotational speed measurement of both gears is shown in Fig. 10. A linear relationship between losses and speed can be observed for both gears. The losses in the CMG are largely due to electromagnetic losses, which should exhibit a quadratic relation with the speed (frequency). The higher than expected linear no-load losses implies that a large mechanical friction component is likely present in the CMG. Fig. 10Open in figure viewerPowerPoint No-load losses of gears at different speeds (high-speed side) 6.3 Load tests The load tests were performed at stepped loads from 25 to 132 N m torque on the LS shaft for a speed range from 80 to 160 rpm. The same tests were performed on both the mechanical and CMG at temperatures of and again at an artificially rendered ambient conditions. The latter is the typical on-site ambient temperature of ACC systems. Figs. 9a and b are the photos of the test bench set-up for mechanical and magnetic gears, respectively. The measured input and output powers of the gears were measured with Lorenz torque sensors. The resultant power losses and efficiencies for the mechanical gear in down-speed configuration are tabulated in Tables 8 and 9, respectively. The efficiency of the mechanical gear ranges from 78 to 79% at low load conditions to above 95% at rated loads. It can be observed that the losses increase with both load and speed for mechanical gears. Table 8. Mechanical gear power losses (W) for different speeds and torques Speed, rpm Torque, N m 25 50 75 100 132 160 81.6 76.4 88.8 102.6 102.5 140 67.9 78.6 83.2 86.8 70.6 120 61.0 63.0 69.9 73.8 72.9 100 51.5 56.0 56.8 61.9 58.8 80 36.2 42.3 40.7 41.3 45.6 Table 9. Mechanical gear efficiency (%) for different speeds and torques Speed, rpm Torque, N m 25 50 75 100 132 160 78.9 91.3 92.7 94.1 95.2 140 79.0 90.2 92.7 94.0 96.1 120 83.1 90.4 92.7 94.0 95.3 100 77.9 89.4 93.0 94.1 95.4 80 78.9 90.7 93.4 94.9 95.6 The resultant power losses and efficiencies for the CMG in down-speed configuration are tabulated in Tables 10 and 11, respectively. The efficiency of the magnetic gear ranges from 72 to 75% at low load conditions to above 93% at rated loads. It is evident that the mechanical gear losses are closely related to both the load torque and speed while the CMG losses are mostly related to rotational speeds. During the test, especially at no-load, a slight eccentricity of the output shaft was observed, which caused additional system vibration and noise. The CMG operated smoother and quietly with increasing load torque at most of the test speeds, which may explain why the losses slightly reduce with increasing load. At a speed of 140 rpm, the system experiences a resonance (∼97.66 Hz) and its vibration and noise intensifies with the load torque increase. Table 10. Magnetic gear power losses (W) for different speeds and torques Speed, rpm Torque, N m 25 50 75 100 132 160 161.8 159.8 158.3 156.0 155.1 140 133.9 134.8 137.9 148.8 153.0 120 116.1 115.5 112.3 110.8 110.6 100 95.3 92.8 91.5 95.6 91.0 80 74.7 76.0 75.1 73.4 73.7 Table 11. Magnetic gear efficiency (%) for different speeds and torques Speed, rpm Torque, N m 25 50 75 100 132 160 75.0 85.0 88.6 91.7 93.5 140 72.2 85.2 88.6 91.0 92.7 120 75.8 85.1 89.7 91.9 93.8 100 74.4 84.6 89.9 91.7 93.9 80 75.7 84.5 89.7 92.0 94.0 6.4 Peak torque measurement of CMG The peak torque measurement of the CMG requires the gear shafts to be slipped out of synchronisation. With the HS shaft of the gear clamped in a frame, the LS shaft was connected to an adjustable steel arm. The LS shaft was then slowly rotated until the shafts slip out of the magnetic pole coupling. In Fig. 11, the measured 2D and 3D FE simulated peak torque values are plotted. As expected the 2D simulation results are the highest at 294 N m. The 3D simulation results are much closer to the measured at 254 N m. The measured peak torque results achieved a value of 243 N m. This can be attributed to manufacturing tolerances and unaccounted flux leakage in the gear. Fig. 11Open in figure viewerPowerPoint Peak torque measurement with 2D and 3D FE simulation values as reference 7 Performance comparison Apart from the high than expected mechanical loss identified on the testing of the CMG, the gear performed generally well. The efficiency across all tests on average fell within 2% lower compared to the mechanical gear. Temperature has relatively little influence on the performance of the mechanical gear and the magnetic gear at rated conditions. The magnetic gear merely experienced a slight decrease in performance during rated conditions at higher temperatures. Both gear performed similarly as a speed increaser and a speed reducer, respectively. Table 12 summarises the measured efficiencies () of both gears under different operating conditions and their respective volumetric and mass torque densities. Table 12. Performance summary of the both gears Type of test Mechanical gear CMG : (speed reducer, ) 95.2% 93.5% : (speed reducer, ) 94.9% 93.7% : (speed increaser, ) 95.7% 92.8% : (speed increaser, ) 95.2% 92.9% : (reducer, 1.5 pu torque) 96.0% 94.9% : (increaser, 1.5 pu torque) 95.7% 95.0% total mass, kg 52 21.7 volume (excl. shafts), 12.3 3.56 mass torque density, N m/kg 2.54 6.08 volume torque density, kN m/m3 10.74 37.03 To analyse the possible cause of the excessive mechanical losses of the CMG, the air-gap forces of the magnetic gear were analysed using Maxwell stress tensor method as described in [25]. For a given gear position the computed distributed radial and tangential force components were first converted to 2D Cartesian components and then integrated along the air-gap contour to find the resultant force. For the flux modulator, the resultant forces from both inner and outer air-gaps are summed and then converted back to radial and tangential components. Fig. 12 shows the magnitude of the unbalanced magnetic forces (UMFs) (in radial direction) exerted on the flux modulator as a function of the angular position of the HS rotor. The magnitude of the average resultant UMF was found to be 167 N. These UMFs can significantly increase bearing friction losses, which may explain the higher mechanical losses of the CMG. Fig. 12Open in figure viewerPowerPoint Magnitude of UMFs (in radial direction) on the flux modulator as a function of HS rotor position at rated load condition 8 Conclusion In this paper, the design optimisation and performance evaluation of a CMG prototype is presented. The CMG is designed, manufactured and experimentally assessed against an equivalent commercial mechanical gear. The CMG performed well in all the tests with the mechanical gear leading by a small margin in most cases. Under rated load conditions, the mechanical gear achieved a peak measured efficiency of 95.2% closely followed by the CMG measuring 93.5%. Both gears achieved similar performance results in up- and down-speed configuration as well as at ambient temperatures of and . The UMFs in the CMG is likely the cause for the high bearing losses. A judicious selection of number of PM poles and modulator pole-pieces is essential for mitigating this adverse effect and realising a high-performance CMG design. With the potential of low maintenance, noise and vibration as well as overload protection, the CMG is a viable alternative for the mechanical gear for ACC applications. Future work will investigate the dynamic behaviour of the CMG for ACC applications. 9 Acknowledgment This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement no. 654443. The authors thank Dr Stiaan Gerber for his help with creating the computer script for calculating the unbalanced magnetic forces. 10 References 1Wurtz, W.: ' Air-cooled condensers eliminate plant water use'. Power, September 2008, Available at http://www.powermag.com, accessed 1 May 2017 2Shaltout, A.A.: 'Analysis of torsional torques in starting of large squirrel cage induction motors', IEEE Trans. Energy Convers., 1994, 9, (1), pp. 135– 142 3Li, Y., Tang, D., Zhang, J., et al: ' An experimental study of pulsatile airflow in the fan inlet of an air-cooling heat exchanger using power spectrum analysis'. Proc. IEEE Power Engineering & Automation Conf., Wuhan, China, 2011, pp. 138– 141 4Zhao, W., Liu, P.: ' Effect of wind on recirculation of direct air-cooled condenser for a large power plant'. Proc. Asia-Pacific Power & Energy Engineering Conf., Wuhan, China, 2009, pp. 1– 4 5Goldshagg, H.: ' ACC fan gearboxes: Eskom's experience in selection and maintenance of ACC gearboxes'. ACCUG Nevada Conf., Nevada, 2013 6Parey, A., Jain, N.K., Koria, S.C.: 'Failure analysis of air cooled condenser gearbox', Case Stud. Eng. Failure Anal., 2014, 2, (2), pp. 150– 156 7Atallah, K., Howe, D.: 'A novel high-performance magnetic gear', IEEE Trans. Magn., 2001, 37, (4), pp. 2844– 2846 8Rasmussen, P., Andersen, T., Jorgensen, F., et al: 'Development of a high performance magnetic gear', IEEE Trans. Ind. Appl., 2005, 41, (3), pp. 764– 770 9Tlali, P.M., Wang, R-J., Gerber, S.: ' Magnetic gear technologies: a review'. Proc. Int. Conf. Electrical Machines, Berlin, Germany, September 2014, pp. 544– 550 10Gouda, E., Mezani, S., Baghli, L., et al: 'Comparative study between mechanical and magnetic planetary gears', IEEE Trans. Magn., 2011, 47, (2), pp. 439– 450 11Nakamura, K., Fukuoka, M., Ichinokura, O.: 'Performance improvement of magnetic gear and efficiency comparison with conventional mechanical gear', J. Appl. Phys., 2014, 115, p. 17A314 12Atallah, K., Calverley, S., Howe, D.: 'Design, analysis and realisation of a high-performance magnetic gear', IEE Proc. Electric Power Appl., 2004, 151, (2), pp. 135– 143 13Gerber, S., Wang, R-J.: 'Design and evaluation of a magnetically geared PM machine', IEEE Trans. Magn., 2015, 51, (8), p. 8107010 14Gerber, S., Wang, R-J.: ' Analysis of the end-effects in magnetic gears and magnetically geared machines'. Proc. Int. Conf. Electrical Machines, Berlin, Germany, September 2014, pp. 396– 402 15Frank, N., Toliyat, H.: 'Analysis of the concentric planetary magnetic gear with strengthened stator and interior permanent magnet inner rotor', IEEE Trans. Ind. Appl., 2011, 47, (4), pp. 1652– 1660 16Gerber, S., Wang, R-J.: ' Evaluation of a prototype magnetic gear'. Proc. IEEE Int. Conf. Industrial Technology, Cape Town, South Africa, February 2013, pp. 319– 324 17Zanis, R., Borisavljevic, A., Jansen, J., et al: ' Modeling, design and experimental validation of a small-sized magnetic gear'. Int. Conf. Electrical Machines and Systems, Busan, Korea, 2013, pp. 560– 565 18Uppalapati, K., Bomela, W., Bird, J., et al: ' Construction of a low speed flux focusing magnetic gear'. IEEE Energy Conversion Congress and Exposition, Denver, CO, USA, 2013, pp. 2178– 2184 19Fernando, N., Saha, S.: 'Torsional shear stress minimization techniques and implications on electromagnetic performance of flux-modulated double rotors', IEEE Trans. Energy Convers., 2018, 33, (1), pp. 49– 58 20Budynas, R.G., Nisbett, J.K.: ' Shigley's mechanical engineering design' ( McGraw-Hill, New York, 2008, 9th edn.) 21 TUFNOL Composites Limited: ' Thermoplastic materials for engineering components', 2003, Available at: http://www.tufnol.com/materials-full/thermoplastic-full.aspx, accessed 1 September 2017 22Ruoho, S., IKäheimo, J., Arkkio, A.: 'Interdependence of demagnetization, loading, and temperature rise in a permanent-magnet synchronous motor', IEEE Trans. Magn., 2010, 46, (3), pp. 949– 953 23Casadei, D., Mengoni, M., Serra, G., et al: ' Behavior of a five-phase surface-mounted permanent magnet motor under magnet demagnetization'. IEEE Int. Symp. Diagnostics for Electric Machines, Power Electronics & Drives, Bologna, Italy, 2011, pp. 265– 271 24Casadei, D., Filippetti, F., Mengoni, M., et al: ' Detection of magnet demagnetization in five-phase surface-mounted permanent magnet generators'. IEEE Int. Symp. Power Electronics for Distributed Generation Systems, Aalborg, Denmark, 2012, pp. 841– 848 25Rahideh, A., Vahaj, A.A., Mardaneh, M., et al: 'Two-dimensional analytical investigation of the parameters and the effects of magnetization patterns on the performance of coaxial magnetic gears', IET Electr. Syst. Transp., 2017, 7, (3), pp. 230– 245 Citing Literature Volume12, Issue5May 2018Pages 677-683 FiguresReferencesRelatedInformation
Referência(s)