France



I. Introduction&lt;br/&gt;Permanent magnet (PM) motors have been widely used in electric vehicles due to its high torque/power density and high efficiency. Yet, a relatively constant airgap magnetic field of PM motors may limit the speed range to large extent, which is still the shortfall for the further applications in EVs. In recent years, a type of variable-flux-leakage PM motor is emerged, which draws intense interest from academics of motor field because of the achievement of flux-controllable capability [1]. In this type of motors, the leakage flux can be varied by designing flux bridge and magnetic barrier, and it can be controllable through the change of &lt;i&gt;q&lt;/i&gt;-axis current of winding. Compared with conventional PM motor, the variable flux is essentially derived from active design of flux leakage. So, the magnetic field distribution of leakage flux is sensitively linked to the flux controllability.&lt;br/&gt;Then, in [2], a variable-flux-leakage PM Motor is proposed, where a unique flux leakage path is purposely considered by arranging the PM position and magnetic barrier reasonably. The results indicate that the relatively obvious capability of flux regulation is achieved by effectively controlling the leakage flux. Yet, in flux-regulated process, the magnetoresistance characteristic of the motor is changed appreciably caused by the variation of leakage flux. It means that different flux-leakage operation modes of the motor will bring the characteristic differences of output torque. For instance, the varied main flux caused by leakage flux greatly leads to the change of PM torque, while the variation of inductance characteristics induced by magnetic saturation bring about the change of reluctance torque. Therefore, to the variable-flux-leakage PM motor, it is necessary and meaningful to comprehensively consider the torque characteristics at various flux-leakage operation modes. And, in this case, how to realize the optimal distribution of different torque components is full of challenge.&lt;br/&gt;In this paper, a flux-controllable permanent magnet (FC-PM) motor is proposed. The key of the study is to conduct torque characteristics analysis and optimization by combining with typical operation conditions. Then, based on different flux-leakage operation modes, a reasonable optimization strategy considering torque components is developed. The results reveal that, based on optimization, the comprehensive torque is effectively improved while maintaining a desirable flux regulation region.&lt;br/&gt;II. Motor Topology and Simulation Result&lt;br/&gt;Fig.1(a) shows the structure of the FC-PM motor with independent separated flux leakage branches. It can be seen that the &lt;i&gt;q&lt;/i&gt;-axis is equipped with multi-layer arc and ellipse magnetic barriers to obtains good magnetic flux leakage controllable performance, which further broadens the motor speed regulation range. Fig.1(b) shows the magnetic density distributions of the FC-IPM motor under different flux-leakage operation modes, and the corresponding requirements of torque components distribution are listed. In Fig. 2(a) and (b), the sensitivity of each design variable to different design objectives is evaluated, and three variables with high sensitivity are selected for further optimization. Fig. 2(c) and (d) show the result of torque distribution under different flux-leakage Operation Modes, which can be seen that through the multi-objective optimization proposed in this paper, the torque quality is improved as expect. The torque performances and corresponding torque component distribution under different modes are shown in Fig.2 (d) and (f). The prototype of proposed FC-IPM motor and corresponding experiment result of back-EMF are shown in Fig. 2(g) and Fig. 2(h), respectively. In additional,&lt;br/&gt;III. Conclusion&lt;br/&gt;In this paper, a FC-PM motor is purposely designed, and a multi-objective optimization method is also newly proposed, where the optimal distribution of torque components under different flux-leakage operation modes is considered. Through optimization, the PM torque becomes the main component under less flux leakage, and the reluctance torque as the main component under more flux leakage. It means that the optimal proportion of different torque components of the FC-PM motor is realized. Both the simulation and experiment results not only verify the reasonability of the proposed motor but also confirm the effectiveness of the proposed optimization method. The more electromagnetic performance analysis and experimental results will be provided in the full paper.

**References:**

[1] X. Zhu, W. Wu, S. Yang, Z. Xiang and L. Quan, in &lt;i&gt;IEEE Transactions on Energy Conversion&lt;/i&gt;, vol. 33, no. 4, pp. 2260-2269, Dec. 2018.&lt;br/&gt;[2] T. Kato, M. Minowa, H. Hijikata, K. Akatsu and R. D. Lorenz, in IEEE Transactions on Industry Applications, vol. 51, no. 5, pp. 3811-3821, Sept.-Oct. 2015.

**Images:**
![](https://assets.underline.io/uploads/markdown_image/1/image/374892bf3a14e058bbeaa3286988bcda.jpg)

![](https://assets.underline.io/uploads/markdown_image/1/image/4178320d508adb7f7bdb56859b57fb75.jpg)




INTERMAG 2021

 Torque Characteristics Investigation of A Flux-controllable Permanent Magnet Motor Considering Different Flux-leakage Operation Modes

#### Welcome to the 67th Annual Conference on Magnetism and Magnetic Materials, MMM 2022!

The Conference was held from October 31 to November 4 and offers both in-person and on-demand content.

MMM 2022 is sponsored jointly by AIP Publishing and the IEEE Magnetics Society. Members of the international scientific and engineering communities interested in recent developments in fundamental and applied magnetism are invited to attend and contribute to the technical sessions. The technical program will include invited and contributed papers in oral and poster sessions, invited symposia, a tutorial, and several special sessions. This Conference provides an outstanding opportunity for worldwide participants to meet their colleagues and collaborators and discuss developments in all areas of magnetism research.

<iframe width="560" height="315" src="https://s3.amazonaws.com/pf-user-files-01/u-59356/uploads/2022-11-02/un13zp7/MMM%20Welcome%20Video%20v5.mp4" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

For tickets to the event, please register at the following link: https://magnetism.org/registration

Please register for the MMM 2022 conference to join the event.

MMM 2022

This Conference provides an outstanding opportunity for worldwide participants to meet their colleagues and collaborators and discuss developments in all areas of magnetism research.

I. Introduction Permanent magnet (PM) motors have been widely used in electric vehicles due to its high torque/power density and high efficiency. Yet, a relatively constant airgap magnetic field of PM motors may limit the speed range to large extent, which is still the shortfall for the further applications in EVs. In recent years, a type of variable-flux-leakage PM motor is emerged, which draws intense interest from academics of motor field because of the achievement of flux-controllable capability [1]. In this type of motors, the leakage flux can be varied by designing flux bridge and magnetic barrier, and it can be controllable through the change of q-axis current of winding. Compared with conventional PM motor, the variable flux is essentially derived from active design of flux leakage. So, the magnetic field distribution of leakage flux is sensitively linked to the flux controllability. Then, in [2], a variable-flux-leakage PM Motor is proposed, where a unique flux leakage path is purposely considered by arranging the PM position and magnetic barrier reasonably. The results indicate that the relatively obvious capability of flux regulation is achieved by effectively controlling the leakage flux. Yet, in flux-regulated process, the magnetoresistance characteristic of the motor is changed appreciably caused by the variation of leakage flux. It means that different flux-leakage operation modes of the motor will bring the characteristic differences of output torque. For instance, the varied main flux caused by leakage flux greatly leads to the change of PM torque, while the variation of inductance characteristics induced by magnetic saturation bring about the change of reluctance torque. Therefore, to the variable-flux-leakage PM motor, it is necessary and meaningful to comprehensively consider the torque characteristics at various flux-leakage operation modes. And, in this case, how to realize the optimal distribution of different torque components is full of challenge. In this paper, a flux-controllable permanent magnet (FC-PM) motor is proposed. The key of the study is to conduct torque characteristics analysis and optimization by combining with typical operation conditions. Then, based on different flux-leakage operation modes, a reasonable optimization strategy considering torque components is developed. The results reveal that, based on optimization, the comprehensive torque is effectively improved while maintaining a desirable flux regulation region. II. Motor Topology and Simulation Result Fig.1(a) shows the structure of the FC-PM motor with independent separated flux leakage branches. It can be seen that the q-axis is equipped with multi-layer arc and ellipse magnetic barriers to obtains good magnetic flux leakage controllable performance, which further broadens the motor speed regulation range. Fig.1(b) shows the magnetic density distributions of the FC-IPM motor under different flux-leakage operation modes, and the corresponding requirements of torque components distribution are listed. In Fig. 2(a) and (b), the sensitivity of each design variable to different design objectives is evaluated, and three variables with high sensitivity are selected for further optimization. Fig. 2(c) and (d) show the result of torque distribution under different flux-leakage Operation Modes, which can be seen that through the multi-objective optimization proposed in this paper, the torque quality is improved as expect. The torque performances and corresponding torque component distribution under different modes are shown in Fig.2 (d) and (f). The prototype of proposed FC-IPM motor and corresponding experiment result of back-EMF are shown in Fig. 2(g) and Fig. 2(h), respectively. In additional, III. Conclusion In this paper, a FC-PM motor is purposely designed, and a multi-objective optimization method is also newly proposed, where the optimal distribution of torque components under different flux-leakage operation modes is considered. Through optimization, the PM torque becomes the main component under less flux leakage, and the reluctance torque as the main component under more flux leakage. It means that the optimal proportion of different torque components of the FC-PM motor is realized. Both the simulation and experiment results not only verify the reasonability of the proposed motor but also confirm the effectiveness of the proposed optimization method. The more electromagnetic performance analysis and experimental results will be provided in the full paper.

**References:**

[1] X. Zhu, W. Wu, S. Yang, Z. Xiang and L. Quan, in IEEE Transactions on Energy Conversion, vol. 33, no. 4, pp. 2260-2269, Dec. 2018. [2] T. Kato, M. Minowa, H. Hijikata, K. Akatsu and R. D. Lorenz, in IEEE Transactions on Industry Applications, vol. 51, no. 5, pp. 3811-3821, Sept.-Oct. 2015.

**Images:**
![](https://assets.underline.io/uploads/markdown_image/1/image/374892bf3a14e058bbeaa3286988bcda.jpg)

![](https://assets.underline.io/uploads/markdown_image/1/image/4178320d508adb7f7bdb56859b57fb75.jpg)

poster

INTERMAG 2021 Welcome Video



	
<figure class="video_container">
 <iframe src="https://screencast-o-matic.com/player/crflqFVnUpZ?v=6&ff=1&title=0&controls=1" width=640 height=360 frameborder="0" allowfullscreen="true"> </iframe>
</figure>

Want to buy a ticket? 

Register for INTERMAG 2021 here

[![](https://assets.underline.io/uploads/markdown_image/1/image/17ce76e7960db633258024eef8d6da7f.png)](https://intermag.org/register)

Got your ticket, but still can’t access the event?

Contact us and our team will look into it.

Harvesting of the ambient radio-frequency (RF) energy is of great interest to address the increasing energy demand for emerging applications, e.g., the internet of things, wearable devices, various sensors and implants, and 3D integrated circuits. To use the weak ambient RF power density for such applications, we need rectifiers capable of operating much below 1 μW RF power with high efficiencies. However, the development of such technologies is severely limited by the conventional semiconductor rectifiers [1], which fail to operate in the weak RF limit. We propose a new rectifier concept [2], simultaneously utilizing the spin-orbit torque (SOT) and the Hall effect, that can provide ~200 μV DC voltage from a 500 nW of radio-frequency (RF) power using existing materials, with a power conversion efficiency as high as 71%. The DC voltage strength can be efficiently enhanced to ~300 mV from the same RF power with a series array of such devices while matching the low impedance of the receiver antenna. We estimate a very high curvature coefficient of the proposed rectifier, in the order of 104 V-1, where the theoretical limit of conventional Schottky diodes is determined by the inverse of the thermal voltage q/kBT ~ 40 V-1 [3]. Various SOT materials are being extensively studied for efficient current-induced switching of scaled ferromagnets for memory applications (see, e.g., Ref. [4] and references therein). Here, we propose a new application of SOT in the rectification of weak RF by coupling it to conventional Hall devices. The basic idea can be explained using a Hall bar and a solenoid connected in parallel, see figure. A current-carrying conductor placed in a magnetic field (B) exhibits a voltage drop in the direction orthogonal to both the current and the B-field, known as the Hall effect. If a fraction of the current flows through a solenoid underneath the Hall bar to generate the B-field, the Hall voltage will be unidirectional irrespective of the current direction, leading to a rectification. We propose a structure where such a solenoid is replaced with a bilayer consisting of a material exhibiting SOT and a ferromagnet (FM) with a low anisotropy energy barrier. We design the device such that the current equally divide between the Hall and the SOT layers. The FM magnetization, on average, follows the fraction of the current flowing in the SOT layer, provided that the current can produce sufficiently strong SOT. The FM applies an effective B-field to the Hall layer and thus leads to a similar rectification case as described in the solenoid case. The Hall effect and spin-orbit-torque are both proportional to current density, which improves inversely with device cross-sectional area, providing the largest signals at the nanoscale. To make the proposed rectifier sensitive to weak RF current, we need a material with a high SOT factor, an FM with low anisotropy energy, and a low Gilbert damping constant. An FM with low anisotropy energy is typically achieved by lowering the total magnetic moment or by tuning the FM thickness to optimize near the transition point between in-plane and perpendicular anisotropies or by using isotropic geometries. We consider a soft ferrite with a very low anisotropy field and a low total magnetic moment, that is expected to easily follow a SOT induced by a weak RF current. Such a low anisotropy energy magnets can achieve a wide frequency bandwidth of operation, depending on the total magnetic moment in the FM and the angular momentum conservation [5]. We simulate the magnetization dynamics using the stochastic-Landau Lifshitz Gilbert equation, which considers the stochastic nature of such low anisotropy energy magnets due to the presence of thermal noise. We can further optimize the device structure and use emerging materials to enhance the figure-of-merits at even lower RF power. The proposed device can lead to important new technologies addressing the alarming energy issues in the era of the internet of things, wearable devices, and densely integrated 3D circuits. This work is supported by the Center for Energy Efficient Electronics Science (E3S), NSF Award 0939514.

**References:**

[1] S. Hemour et al. IEEE Trans. Microw. Theory Tech. 62, 965–976 (2014). [2] S. Sayed, S. Salahuddin, and E. Yablonovitch, to appear in Appl. Phys. Lett. (2021). [3] J. Karlovský, Solid-State Electronics 10, 1109 – 1111 (1967). [4] S. Sayed et al., arXiv:2010.01484 (2020). [5] S. Sayed et al., Phys. Rev. Applied11,054063 (2019).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/e204cbc47f74316fbbb8a9936ad0cad3.jpg)
(a) Basic concept of rectification explained using a Hall bar and a solenoid sharing the same current. The magnetic field on the Hall bar, applied by the solenoid reverses when we reverse the current direction. Thus the Hall voltage remains of the same polarity irrespective of the current direction. (b) The proposed spin-orbit torque (SOT) rectifier, where the solenoid is replaced with a bilayer of a SOT material and a soft ferromagnet (FM). The FM follows in the current in the SOT material and applies a magnetic field to the Hall layer, similar to the solenoid case.

 Spin-orbit torque rectifier for energy harvesting from weak radio-frequency

The ongoing development in the landscape of low power electronics has paved the way for the emergence of autonomous wireless sensor networks (WSNs) to create a smart and responsive environment around us through the Internet of Things (IoT). For instance, efficient WSN performs structural health monitoring of heavy-duty vehicles like train and reports on vital physical parameters and alerts if there are any probable failure to reduce the chances of accidents. Perpetually powering these sensors is one of the critical bottlenecks in realizing a WSN system. The associated maintenance and installation cost [1] of the batteries limits the widespread deployment of the WSNs. The demand for the implementation of battery-less microsystems and the ubiquity of ambient vibrations in an industrial and domestic environment has made the electromagnetic vibration energy harvesters (EM-VEHs) [2] a striking sustainable energy source for powering the WSNs. Compared with the conventionally used narrowband linear VEHs, the nonlinear counterparts [3,4] are much more efficient in harnessing mechanical energy from real-world broadband vibrations owing to their intrinsic wideband frequency response. An effective way of realizing nonlinearity is by tailoring the stiffness of the spring geometry [5.6], which on the other hand depends critically on the thickness of the spring material limiting the scope of geometric alterations. Tapered spring structures on the other hand open up the scope for optimizing the degree of nonlinearity and hence the bandwidth of the VEHs by changing the taper ratio of the spring structure [7]. In this work, we exploit exponential and parabolic tapering to induce stretching based nonlinearity in the spring structure and we explore the overall performance of an optimized tapered EM-VEH that could be a potential candidate for powering WSN involved with railway infrastructure monitoring system. The EM transducer under consideration works on the Faraday’s principle, in which the relative motion between a pair of coil and magnet induces voltage across the coil, which could be harvested in the form of usable electrical energy. Among the other conventional transducers, owing to the reliability, longevity and large tunable parameters offered, the EM transducers are the most promising ones. We have used the structural mechanics module of COMSOL Multiphysics to explore the effect of tapering on the force-displacement relationship of the spring structures. The broad arm of the spring (length w1) is fixed with the frame and the shorter arm (w2) is attached to the moving load, which is the magnet assembly as shown in Figure 2. We have varied the taper ratio (= w2/w1) of the spring structures from 0.2 to 0.35, keeping the length of the shorter arm fixed to 4mm. Figure 1 depicts the first eigenmodes of the structures exhibiting out of plane displacements. Owing to the atypical stress distribution of these tapered structures, at a larger amplitude of excitation, the vertical displacement of the spring varies nonlinearly with the force. Hence, the restoring force of the spring consists of the linear part responsible for bending of the springs and the nonlinear part responsible for the stretching of the same. As compared with the parabolically tapered spring architectures, the exponentially tapered counterpart exhibits larger displacements for the same taper ratio. This can be attributed to the larger stress distribution on the springs owing to the exponentially reducing area. We further designed a nonlinear EM-VEH employing exponentially tapered spring having the taper ratio of 0.25 as shown in figure 2. The central paddle of the spring holds the magnet assembly comprising of four NdFeB magnets arranged such that it forms a close loop of magnetic field lines across the coil. To further eliminate the fringing of magnetic field lines and enhance the EM-coupling, block of soft magnetic steel is arranged on both sides of the magnet assembly, resulting in an EM-coupling of 15.2Wb/m. At low acceleration (0.1g), the system’s frequency response resembles with that of a linear VEH, the load power maximizing at 124Hz. As the excitation amplitude increases from 0.1g to 1g, the frequency response becomes gradually non-resonant in nature. When the frequency is swept in the forward direction, the load power changes from 235μW at 124Hz to 1.6mW at 154Hz with 1g acceleration, beyond this point the response suddenly falls off the high energy branch, which is a signature of the effect of induced nonlinearity. But when the frequency is swept in the backward direction, the response follows a lower energy branch, giving rise to the phenomena of multistability, which in turn aids in enhancing the operable bandwidth of frequency that the device offers. At 1g acceleration, the system offers 20Hz of bandwidth. The typical vertical acceleration of a diesel train is around 0.93g [8] with substantial mechanical energy distributed around 100-300Hz (Fig.2d). This developed VEH could be employed to harness this mechanical energy from trains to power the WSNs associated with the railway system responsible for structural health monitoring.

**References:**

[1] J. P. Amaro, R. Cortesão, J. Landeck et al, IEEE Transactions on Instrumentation and Measurement, vol. 64, no. 7, pp. 1847-1857 [2] W. Hwang et al., Applied Energy, vol. 243, pp. 313-320, 2019. [3] S. Roy, P. Podder, and D. Mallick, IEEE Magnetics Letters, vol. 7, pp. 1-4, 2016. [4] P. Constantinou and S. Roy, Smart Materials and Structures, vol. 25, no. 9, p. 095053, 2016. [5] B. Marinkovic and H. Koser, Applied Physics Letters, vol. 94, 2009. [6] D. Mallick, A. Amann, and S. Roy, Smart Materials and Structures, vol. 24, no. 1, p. 015013, 2014. [7] K. Paul, A. Amann, S. Roy, Applied Energy, 2020, 116267, ISSN 0306-2619. [8] Real vibration, NIPS lab, https://realvibrations.nipslab.org/node/1093.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/211fa84d35c6420218da27712abd9a60.jpg)
Fig 1: First eigenmodes of parabolically (a) and exponentially (b) tapered spring structure with different taper ratio is shown. The force-displacement relation of springs is shown in (c).

![](https://assets.underline.io/uploads/markdown_image/1/image/5a7acb6abd7048009ffd86b9d85325f7.jpg)
Fig 2: Exponentially tapered EM-VEH (a) is designed with taper ratio 0.25. The EM interaction is depicted in (b). Frequency response of the nonlinear VEH is shown in (c) and power spectral density of train vibrations in (d).

 Design optimization of tapered wideband nonlinear electromagnetic vibration energy harvesters for powering IoT

This paper analyzes the influence of harmonic back-EMFs, especially the third harmonic back-EMF, on the flux-weakening performance of interior permanent magnet synchronous motor (IPMSM). First, IPMSM model and the back-EMF harmonics components of IPMSM are studied. Then, the commonly used flux-weakening control methods are introduced briefly. Finally, the influence of back-EMF harmonics on the flux-weakening performance is analyzed in detail from the perspective of voltage limit. The analysis is validated with the simulation. 1. Introduction IPMSM has been widely explored due to its excellent performance such as high efficiency and high power density. In many applications of IPMSM such as electric vehicles, a wider speed range is usually required, and flux-weakening control is usually adopted to obtain higher speeds. The flux-weakening range of permanent magnet motor is very limited, because the magnetic field of the permanent magnet cannot be adjusted. Generally speaking, IPMSM has a wider flux-weakening speed range than surface permanent magnet synchronous motor (SPMSM) because of the salient-pole structure [1]. Based on the current vector control, there are two typical flux-weakening control methods [2]: one is the feedforward control method, which requires the accurate IPMSM mathematical model parameters. The other is the feedback control method, which uses the PI regulator to compensate the demagnetizing current. The feedback control method can be further divided into two categories: flux-weakening control based on dual current regulators and flux-weakening control based on variable voltage angle. The dual current regulators method is still based on current vector control, but suffers from the conflict of two voltages saturation. There is no current regulator in the initial variable voltage angle method. However, the single d-axis current regulator or single q-axis current regulator methods are developed to plan the current locus and improve the dynamic performance [3]. Regardless of the different flux-weakening methods, its essence is to offset the permanent magnet back-EMF at high speeds by applying the demagnetizing current, and relieve the conflict of back-EMF and inverter voltage limit. However, the demagnetizing current can only counteract the fundamental back-EMF. The back-EMF of IPMSM has rich harmonic back-EMFs. The reason is that the permanent magnet flux linkage contains the harmonic components. Among these harmonics, the third harmonic is the most obvious. At high speeds, the high third harmonic back-EMF can also reach the inverter voltage limit like the fundamental back-EMF. However, there is no demagnetizing current offsetting it. This paper will analyze the influence of the harmonics back-EMFs on the flux-weakening performance of IPMSM. The rest of the paper is organized as follows: Section 2 introduces the IPMSM model and harmonics back-EMFs; Section 3 gives a profile of existing flux-weakening method; Section 4 analyzes the influence of harmonics back-EMFs on the flux-weakening performance of IPMSM. 2. IPMSM Model and back-EMF harmonics Flux-weakening control is based on the current vector control architecture, which is implemented in synchronous rotating (d-, q-axis) coordinate system. Through coordinate transformation, all variables in the IPMSM mathematical model, including the back-EMF, become DC components. However, this mathematical model only considers the fundamental back-EMF and does not consider the harmonic components in the back-EMF. In fact, due to the permanent magnet shape and the magnetization direction, the back-EMF contains rich harmonic components. Especially for IPMSM, since its permanent magnets are embedded in the rotor, it is more difficult to optimize the permanent magnets shape than the surface mount structure. Therefore, IPMSM contains more harmonics in the back EMF than SPMSM. Fig.1 shows a IPMSM’s back-EMFs. It can be seen that they are not standard sine waves. Through FFT analysis, the THD of back-EMF reaches 10.25%, and the main harmonics are the third harmonic. 3. Dual current regulators based flux-weakening method This paper adopts the commonly used dual current regulators based flux-weakening method. The principle of this method is adding a voltage feedback regulator. When the reference voltage exceeds the inverter voltage limit, a negative demagnetizing current is generated through the PI regulator, and then a voltage that cancels the fundamental back-EMF is generated to eliminate the saturation of the reference voltage. 4. The influence of harmonic back-EMFs on the flux-weakening performance. The essence of flux-weakening control is to counteract the high back-EMF through the voltage generated by the demagnetizing current. However, the demagnetizing current in the d-axis can only offset the fundamental back-EMF. The harmonic back-EMFs are also very large at high speeds, and can also cause saturation of the inverter voltage. Fig.2 shows the back-EMFs of the IPMSM operating in the flux-weakening condition. In the simulation, the inverter voltage limit is set as 150V. It can be seen that through applying d-axis demagnetizing current, the fundamental back-EMF has been suppressed. But the amplitudes of the harmonics back-EMFs reach 130V, which is very close to the inverter voltage limit. In other words, for higher speeds, the harmonics back-EMFs will reach the inverter voltage limit, which will prevent the speed from continuing to rise. This proves that the harmonics back-EMFs will seriously affect the range of flux-weakening control.

**References:**

[1] N. Bedetti, S. Calligaro and R. Petrella, "Analytical Design and Autotuning of Adaptive Flux-Weakening Voltage Regulation Loop in IPMSM Drives With Accurate Torque Regulation," in IEEE Transactions on Industry Applications, vol. 56, no. 1, pp. 301-313, Jan.-Feb. 2020 [2] Wang J, Wu J, Gan C, et al. Comparative study of flux-weakening control methods for PMSM drive over wide speed range[C]. 2016 19th International Conference on Electrical Machines and Systems (ICEMS) 2016: 1-6. [3] Z. Zhang, C. Wang, M. Zhou and X. You, "Flux-Weakening in PMSM Drives: Analysis of Voltage Angle Control and the Single Current Controller Design," in IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 7, no. 1, pp. 437-445, March 2019.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/96257dd369786bc21f178f22f521736d.jpg)

![](https://assets.underline.io/uploads/markdown_image/1/image/e31f7fd4e22a7bac9f291f7041c2cdfc.jpg)

 Research on the influence of harmonics back-EMFs on the flux-weakening performance of IPMSM

I. Introduction The electric vehicle market is expanding, and the demand for components related to electrification of powertrains is increasing rapidly. There is a traction motor as a related key component, and interior permanent magnet synchronous motor (IPMSM) having excellent efficiency and power density is mainly applied [1]. In addition, NdFeB permanent magnet (NdFeB-PM) with high energy product is used for traction IPMSM. However, NdFeB-PM is expensive and there is an issue with stable supply. As a traction motor without NdFeB-PM, there have been representative studies in terms of permanent magnet assisted-synchronous reluctance motor (PMa-SynRM) using ferrite-PM, switched reluctance motor (SRM) and separately excited synchronous motor (SESM) [2-4]. However, PMa-SynRM with ferrite-PM has a lower power density than conventional IPMSM. SRM structurally has a major issue with acoustic noise and SESM has a disadvantage that requires a separate device for generating the field flux of the rotor. Recently, in order to reduce the amount of NdFeB-PM in IPMSM, there is a study on a hybrid-type PM motor (HPMM) that simultaneously applies ferrite-PM to the rotor [5-7]. However, if the HPMM does not have a stator, there is a problem that ferrite-PM become demagnetized due to the strong leakage flux of NdFeB-PM [5-6]. In addition, in order to achieve the same torque density as the target IPMSM, there are problems in that the use of ferrite-PM must be increased by reducing the inner diameter of the rotor [7]. Therefore, in this paper, we solve these problems and propose a new HPMM based on a trapezoidal ferrite-PM that has the same torque density as the traction motor of Nissan Leaf electric vehicle commercialized in 2018, the target motor. At the same time, the proposed HPMM reduces the use of NdFeB-PM by 30% and the total PM cost by 14.8% compared to the target motor. II. Structure of the Proposed HPMM Fig. 1 (a) and (b) show the ¼ rotor cross-sectional view of the proposed HPMM and the target motor. The stator of the two motors is identical, which is the stator of the 2018 Nissan Leaf traction motor. In the proposed motor, NdFeB-PM and ferrite-PM are arranged on the d-axis direction. When the two types of PM are arranged in the d-axis direction, the leakage flux of NdFeB-PM converges by itself or is aligned in the same direction as the magnetization direction of ferrite-PM, so that demagnetization of ferrite-PM does not occur. However, since the permeability of the magnet is almost similar to that of air, this structure generates a high reluctance of the d-axis, resulting in a decrease in magnet torque of IPMSM. To overcome this point, the q-axis flux path is enlarged by applying trapezoidal ferrite-PM. That is, the d-axis inductance decreases as the d-axis reluctance increases, but the q-axis inductance is maintained or increased according to the expansion of the q-axis flux path, thereby improving the reluctance torque. These characteristics are expressed in Fig. 1 (a), the performance of the proposed HPMM and the target motor are predicted by FEM using commercial software, JMAG-Designer and Solidworks under the same operating condition. III. FEA Verification Fig. 1 (c) is the flux density distribution of the rotor in the absence of a stator. The temperature of the PM was selected as -40 degC considering the low-temperature demagnetization of ferrite-PM and the automotive environment temperature range. As described in Chapter II above, it is shown that demagnetization of ferrite-PM is not generated by the leakage flux of NdFeB-PM. The lowest flux density of ferrite-PMs is 0.3 T, which is not less than knee point of 0.05 T. Fig. 1 (d) shows the average torque for each phase angle β of the proposed HPMM and target motor. The input current is 465 A, the maximum current of the target motor, and PMs temperature is 100 degC. In β = 0 deg, which means magnet torque, the magnet torque of the proposed HPMM is 198 Nm, which is 27.5 Nm lower than that of the target motor. However, with the improved reluctance torque of the proposed HPMM, the maximum torque is 336.9 Nm, which is almost the same as the target motor. Fig. 1 (e) shows the results of the mechanical stress analysis of the rotor at a maximum speed of 10,500 rpm. Since the traction motor for an electric vehicle performs variable speed operation control over a wide speed range, it is required to secure a safety factor for rotor rigidity at the maximum speed. In Fig 1 (e), based on the same analysis conditions and material properties, the maximum stresses of the proposed HPMM and the target motor are 305 MPa and 351 MPa. Based on the yield stress of the electrical steel sheet, 420 MPa, the safety factors are 1.38 and 1.2, respectively. Table I shows the specifications of the proposed HPMM and target motor. The proposed HPMM realizes the same torque density as the target motor, and at the same time reduces the usage of NdFeB-PM by 30 % and the total PM cost by 14.8 %. IV. Summary This paper proposes a new HPMM for electric vehicle traction motor. The proposed HPMM solves the problem of existing HPMM and realizes the same torque density as the target motor, Nissan leaf traction motor. At the same time, the proposed HPMM reduces the usage of NdFeB-PM by 30 % and the total PM cost by 14.8 % compared to the target motor.

**References:**

[1] K. T. Chau, et al, IEEE Trans. Ind. Electron., vol. 55, no. 6, pp. 2246–2257 (2008) [2] M. Obata, et al, in Proc. Eur. Conf. Power Electron. Appl., pp. 1–9 (2013) [3] K. Kiyota, et al, IEEE Trans. Ind. Appl., vol. 48, no. 6, pp. 2303–2309 (2012) [4] C. Stancu, et al, IEEE Trans. Ind. Appl., vol. 54, no. 1, pp. 223–232 (2018) [5] C. L. Lim, et al, IEEE Trans. Magn., vol. 53, no. 11, Art. no. 8110904 (2017) [6] C. L. Lim, et al, IEEE Trans. Ind. Appl., vol. 55, no. 5, pp. 4692-4701 (2019) [7] Z. Zhang, IEEE Energy Conversion and Exposition, DOI: 10.1109/ECCE44975.2020.9235708 (2020)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/5db93ba58daa0d53cba74fdc9b2b11fe.jpg)
Fig. 1. Rotor structure and FEA results (a) Proposed HPMM (b) Target motor (c) Magnetic flux density distribution without stator (d) Average torque by current phase angle β (e) Mechanical stress analysis results at rotational speed 10,500rpm

![](https://assets.underline.io/uploads/markdown_image/1/image/a7827f24c41627982c8a3e7c525a9dea.jpg)

 Design of Hybrid-Type PM Motor for Electric Vehicle Traction using Trapezoidal Ferrite-PM to improve Reluctance Torque

I. Introduction Permanent magnet synchronous machines (PMSMs) have been widely used in electric vehicles. However, traditional PMSMs need continuous flux-weakening current in high-speed region, which can reduce the operating efficiency and increase the demagnetization risks of PMs. Variable-flux PMSMs have gained a lot of attention for their ability to expand speed range and reduce loss by manipulating the magnetization state dynamically. Variable-flux machines employ low coercive force magnets, whose magnetization state can be adjusted by d-axis current pulses [1]. The PM excitation field is adjustable and the need for continuous flux-weakening current is reduced. The power density of the variable-flux machine firstly proposed in [1] is relatively low since only AlNiCo is used. The variable-flux PMSMs that employ both AlNiCo and NdFeB are proposed to improve the power density, in which two kinds of magnets are arranged in either series or parallel magnetic circuits [2-4]. Moreover, the magnetic circuit of PM flux is complicated and changeable in hybrid-PM variable-flux PMSMs under different magnetization states, which brings inconvenience to the modelling and analysis of machines. A novel hybrid-PM variable-flux PMSM with series-parallel magnetic circuits is proposed in this paper. The machine topology and operating principle are analysed. The equivalent magnetic circuit model of the machine is established and proven with good precision under different magnetization states. The electromagnetic performances are analysed and compared with the series-connected and parallel-connected variable-flux PMSMs. The results prove that the proposed machine has high torque density, wide magnetization state adjustment range, stable PM working points and high efficiency. II. Machine Topology and Operating Principle The proposed machine employs both AlNiCo and NdFeB, as shown in fig. 1, where red coloured magnets are AlNiCo and blue coloured ones are NdFeB. Two layers of magnets are placed in the proposed machine. In the upper V-shape layer, NdFeB is connected with AlNiCo in series near q-axis, and these two series-connected magnets are connected with another AlNiCo located near d-axis in parallel. In the lower U-shape layer, NdFeB and AlNiCo are connected in series. Under reverse magnetization state, AlNiCo near d-axis in the V-shape layer can be reversely magnetized by negative d-axis current pulses, and most PM flux of the V-shape layer is short-circuited in rotor core. Under forward magnetization state, the flux of U-type magnets flow through the upper V-type magnets and help to stabilize PM working points. The proposed machine combines the advantages of the parallel-connected and series-connected machines. III. Equivalent Magnetic Circuit Model The magnetic circuit in the proposed machine is complicated and changeable since the amplitude and direction of PM flux vary with machine's magnetization state. It's necessary to establish the equivalent magnetic circuit model, which can provide an effective way for the operating principle and performance analysis. The equivalent magnetic circuit model of the proposed machine under different magnetization states is established, with PMs substituted by equivalent flux sources and reluctances, and iron core and air gap replaced by equivalent reluctances, as shown in fig. 1. More details are presented in full paper. The air-gap flux density, back-EMF and PM operating points are obtained by this model, and compared with the FEA results. The calculation errors are below 5%, which proves the accuracy of the proposed equivalent model. The complicated and changeable series-parallel-connected magnetic circuit in the proposed machine under different magnetization states is equivalently simplified by this model, which brings more convenience to the machine modelling and the fast performance analysis and calculation. IV. Performance Analysis The electromagnetic performances, including back-EMF, torque capacity, operating points of AlNiCo, magnetization state adjustment range and operation efficiency, of the proposed machine are analysed and compared with the parallel-connected and series-connected variable-flux machines. The fundamental amplitudes of no-load EMF of these three machines under fully forward magnetized state are 180V, 165V and 187V. And those under fully reversed magnetized state are 60V, 38V and 120V. The magnetization state adjustment ratios, which are the important index for variable-flux PMSMs, are 66.7%, 77.0% and 35.8%. For the parallel-connected machine, the working points of AlNiCo will move below the knee points when the load current angle is less than -37°. AlNiCo is demagnetized unexpectedly then. For the proposed machine, the working points of AlNiCo are stable. The operating loss for the proposed machine under different magnetization states and operation conditions is evaluated, as shown in fig. 2. A low magnetization state is suitable for the high speed, low torque condition; A high magnetization state is suitable for the low speed, high torque condition. The operating efficiency is significantly improved by adopting proper magnetization states under different operating conditions. The performances are analysed in detail in full paper. The results prove that the proposed machine has the characteristics of high torque density, wide magnetization state adjustment range, stable PM working points and high efficiency.

**References:**

[1] V. Ostovic. "Memory Motors," IEEE Ind. Appl. Mag., 9(1): 52-61, 2003. [2] K. Sakai, K. Yuki, Y. Hashiba, et al. "Principle of the Variable-Magnetic-Force Memory Motor," in Proc. Int. Conf. Electr. Mach. Syst., 2009. [3] H. Hua, Z. Q. Zhu, A. Pride, et al. "Comparative Study on Variable Flux Memory Machines with Parallel or Series Hybrid Magnets," IEEE Trans. Ind. Appl., 55(2): 1408-1419, 2019. [4] J. H. Chen, J. Li, R. H. Qu et al. "Magnet-Frozen-Permeability FEA and DC-Biased Measurement for Machine Inductance: Application on a Variable-Flux PM Machine," IEEE Trans. Ind. Electron., 65(6): 4599-4607, 2018.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/1f615647f59d8d904b1984cf04d6a158.jpg)
Fig. 1. Equivalent magnetic circuit of the proposed machine.

![](https://assets.underline.io/uploads/markdown_image/1/image/e2fa510bb2c70a503cf468f3c2493084.jpg)
Fig. 2. Loss evaluation under different magnetization states and load conditions.

 Design, Modelling, and Analysis of a Novel Series-Parallel-Connected Hybrid-PM Variable-Flux PMSM

The Interior Permanent Magnet Synchronous Motor (IPMSM) is mainly used as a driving motor for Electric Vehicles (EVs) due to its advantage for high-power density and wide speed operation range. Because of the characteristics of EVs motor, it is necessary to perform high-speed driving at a small load condition. Thus, IPMSM is controlled by the field weakening control method to meet a wide operation range, which controls the phase angle through the d- and q-axis transformation. However, as the phase angle increases, the air-gap magnetic flux contains more harmonics and thus becomes less controllable, which has the drawback that there is a limit to the operation range. Moreover, in the case of large commercial EVs, high torque is required at low speed for the driving stability and the gradeability. Thus, to perform high torque at low speed, the motor should have a large permanent magnet in the motor. As the field magnetic flux increases, the back EMF also increases, so a larger phase angle is required for high-speed operation. However, the maximum speed that can be driven through the field weakening control is decreased because of the harmonics. For high-speed driving of commercial EVs, an optimal design method is needed to expand the field weakening region. The reason why the existence of a maximum speed limit is the increased harmonic component as the phase angle increases. In this paper, we divided the design part into the rotor and the stator. And then the effect on the field weakening control is analyzed when the design variables are changed. Finally, a motor design method for high-power at low-speed mode with a wide operation range is proposed by reducing harmonics and enhancing the back EMF controllability. To analyze the effects of the harmonics and the field weakening control characteristics as the design variable changed, no-load analysis is performed according to the number of slots per phase per pole. In the load analysis, the variation of line voltage is analyzed according to the current value when the maximum demagnetizing magnetomotive force is applied. Fig.1 represents that the larger the number of slots per phase per pole, the more stable voltage can be controlled. Namely, the voltage can be decreased by varying the current value and phase angle at a specific speed. It means that the limitation of the field weakening region can be possible calculated. Next, the field weakening control characteristic is analyzed according to the variation of the number of conductors per slot. Although the number of conductors per slot can reduce copper loss, it needs to prepare the countermeasure against voltage rising for field weakening control. To compare the effect according to the number of conductors in the field weakening region, the voltage values according to the number of turns are analyzed under the condition given the maximum demagnetizing magnetomotive force. The maximum operating speed considering the voltage limitation is selected. In consequence, fewer conductors in the slot are more advantageous for high-speed operation, but the rising current for voltage control and torque generation must be considered. The characteristic of field weakening control is analyzed by changing the magnet layer and shape, and the harmonic reduction analysis is implemented by the skewed method. Prior to the design of permanent magnet shape, permanent magnet with multi-layer structures, the air-gap magnetic flux is relatively sinusoidal compared to the single-layer. Moreover, the torque generation is flexible by securing the q-axis flux path. In other words, the multi-layer structure is used despite the difficulties of productivity due to this advantage. It is analyzed how the multi-layer structure affects the field weakening region. Additionally, the Total Harmonics Distortion (THD) and reluctance torque are analyzed according to changes in the number of magnet layers. As a result, the higher the number of layers, the lower the THD and the greater the proportion of the use of reluctance torque. It is indicated that the voltage controllability in the field weakening control range is more effective if the rotor flux linkage is small when the maximum demagnetizing magnetomotive force is applied. Depending on the shape of the permanent magnet, the barrier changes simultaneously, and the d- and q-axis flux path also changes. So, the air gap magnetic field distribution and voltage control performance are changed. The performance of voltage controllability according to the shape of the magnet is compared by representing the magnetic flux density in the air gap. In Fig.2, if the flux linkage is small, the voltage controllability in the field weakening control region is more stable when the maximum demagnetizing magnetomotive force is applied. As a result, V-type models demonstrate effectively reducing total flux linkage over Bar type models. Lastly, the skew is applied for optimized design to reduce harmonics. The lowest voltage curve can be investigated in both the skew and the non-skew models for comparison of harmonic reduction using FFT analysis. In this paper, with the main aim of optimizing the characteristics in the field weakening control region and improving the voltage control performance, the design of a motor for driving 300kW commercial EVs with a wide range of field weakening control region is presented.

**References:**

[1]Ki-Chan Kim, "A Novel Magnetic Flux Weakening Method of Permanent Magnet Synchronous Motor for Electric Vehicles", IEEE Transactions on Magnetics, Vol.48, No.11, p.4042-4045 (2012) [2]J. A. Guemes, A. M. Iraolagoitia, and P. Fernandez, “Torque Analysis in Permanent-Magnet Synchronous Motors: A Comparative Study”, IEEE Transactions on Energy Conversion, Vol.26, No.1 p.55-63 (2011) [3]Jung-Min Mun, Yong-Jae Kim, and Sang-Yong Jung, "Design Characteristics of IPMSM With Wide Constant Power Speed Range for EV Traction”, IEEE Transactions on Magnetics, Vol. 53, No.6 (2017) [4]Mohammad Sedigh Toulabi, John Salmon, and Andrew M. Knight, “Concentrated Winding IPM Synchronous Motor Design for Wide Field Weakening Applications”, IEEE Transactions on Industry Application, Vol.53, No.3 p.1892-1900 (2017)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/cb8f3fa65f04b2fffc98850c049c66f4.jpg)
Comparison of the lowest line voltage with different number of slots

![](https://assets.underline.io/uploads/markdown_image/1/image/bd596c5eb22c3a03064e9bf2ac8f85b6.jpg)
The comparison of the total flux linkage between 2-layers V-type and Bar-type model

 A Design of IPMSM High-Power Electric Vehicles with Wide Field Weakening Control Region

I Introduction Permanent magnet synchronous machines (PMSMs) have been widely used in electric vehicle industry for their characteristics of high efficiency and power density. However, the development of PMSMs is limited by the unalterable permanent magnet (PM) excitation field. The flux-weakening control used in PMSMs can bring about high copper loss, low efficiency, and PM demagnetization. Variable-flux PMSMs are considered as a good candidate for EV propulsion because the speed range is wider and the operation efficiency is higher than traditional PMSMs. Variable-flux PMSMs employ low coercive force magnets, whose magnetization state can be adjusted by d-axis current pulses [1]. The PM excitation field is adjustable and the need for continuous flux-weakening current is reduced. Two schemes of novel hybrid-PM variable-flux PMSMs with series-parallel magnetic circuits are proposed in this paper. The performances of these two machines are analysed and compared with series-connected and parallel-connected machines. Moreover, the efficiency characteristic of the proposed machine under different magnetization states and load conditions is evaluated. A control strategy of appropriately combining magnetization state adjustment control and flux-weakening control is proposed, which can obviously improve the efficiency in full speed range, especially in high-speed region, by manipulating the magnetization state. II Machine Topology and Operating Principle Two schemes of variable-flux PMSMs with series-parallel magnetic circuits are proposed, as shown in fig.1, which are the single-layer V-type and two-layer V-U-type machines. The red coloured magnets are AlNiCo and blue coloured ones are NdFeB. As shown in fig. 1(a), the V-shape magnets are placed in the V-type machine. The NdFeB is connected with AlNiCo in series near q-axis, and these two series-connected magnets are connected with another AlNiCo located near d-axis in parallel. As shown in fig. 1(b), two layers of magnets are placed in the V-U-type machine. The series-connected U-shape NdFeB and AlNiCo magnets are placed under the V-shape magnets. The magnetization state of proposed machines can be adjusted by manipulating the magnetization direction of AlNiCo near d-axis in V-shape magnets and the flux amplitude of all AlNiCo. More details about the operating principle are analysed in full paper. III Performance Comparisons Electromagnetic performances, including back-EMF, torque capacity, stability of AlNiCo and magnetization state adjustment range, of the proposed machines are analysed and compared with parallel-connected and series-connected machines. Under fully forward magnetized state, the fundamental amplitudes of no-load EMF of V-type and V-U-type series-parallel-connected, parallel-connected and series-connected machines are 168V, 180V, 165V and 187V respectively. And those under fully reversed magnetized state are 58V, 60V, 38V and 120V. The magnetization ratios are 65.5%, 66.7%, 77.0% and 35.8%. Other performances are also compared in detail in full paper. After the comprehensive comparison, the proposed V-U-type series-parallel-connected machine is selected. The magnetizing/demagnetizing current density of V-U-type machine is 17.5A/mm2, which is smaller than that in [2-4]. The torque density of V-U-type machine is nearly the same with that in [3-4], and larger than that in [2]. The analysis results show that the proposed V-U-type series-parallel-connected machine combines the advantages of parallel-connected and series-connected variable-flux PMSMs. IV Efficiency Improvement through Magnetization State Adjustment The expanded speed range and efficiency improvement in high-speed region are the distinct advantages of the proposed machine. The maximum achievable torque-speed ranges at 4 typical magnetization states are evaluated. The speed range is significantly expanded by manipulating the magnetization state. The operating efficiency of the proposed machine at 4 magnetization states are evaluated. The efficiency can be significantly improved by adopting appropriate magnetization states at different working points. A control strategy to improve efficiency by adjusting magnetization state is proposed. For uncertain working conditions, a steady-state efficiency maximizing control strategy is proposed. The magnetization states whose operating efficiency is highest at every torque-speed working point are selected based on the efficiency characteristic at different magnetization states, as shown in fig. 2. A LUT is formed by the selected magnetization states. A hysteresis integral comparator is proposed to minimize the unnecessary magnetization state adjustment. The steady-state efficiency maximizing control is achieved by choosing magnetization states through the LUT and minimizing unnecessary magnetization state adjustment through the hysteresis integral comparator. For certain working conditions, like WLTC, an efficiency maximizing control strategy considering magnetization state adjustment loss is proposed. Optimal magnetization states at every working point are selected based on Genetic Algorithm. The optimization target is to minimize the whole drive cycle loss including the steady-state loss at each working point and the magnetization state adjustment loss. The results prove that the operation efficiency is significantly improved by adopting the proposed magnetization state adjustment strategy.

**References:**

[1] V. Ostovic. "Memory Motors," IEEE Ind. Appl. Mag., 9(1): 52-61, 2003. [2] N. Limsuwan, T. Kato, K. Akatsu, et al. "Design and Evaluation of a Variable-Flux Flux-Intensifying Interior Permanent-Magnet Machine," IEEE Trans. Ind. Appl., 50(2): 1015-1024, 2014. [3] A. Athavale, K. Sasaki, B. S. Gagas, et al. "Variable Flux Permanent Magnet Synchronous Machine (VF-PMSM) Design Methodologies to Meet Electric Vehicle Traction Requirements with Reduced Losses," IEEE Trans. Ind. Appl., 53(5): 4318-4326, 2017. [4] H. Hua, Z. Q. Zhu, A. Pride, et al. "Comparative Study on Variable Flux Memory Machines with Parallel or Series Hybrid Magnets," IEEE Trans. Ind. Appl., 55(2): 1408-1419, 2019.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/89944429fdc8a6824df6f50b6fd02901.jpg)
Fig.1 Two schemes of the proposed machines.

![](https://assets.underline.io/uploads/markdown_image/1/image/53ce01543de2d432e4a23087dffdad54.jpg)
Fig.2 Selected magnetization states with highest efficiency at each working point.

 Study of a High-Efficiency Series-Parallel-Connected Hybrid-PM Variable-Flux Permanent Magnet Synchronous Machine

I Introduction Variable-flux permanent magnet synchronous machines (PMSMs) have gained a lot of attention for their ability to expand speed range and reduce copper loss by manipulating the magnetization state dynamically. Variable-flux PMSMs employ low coercive force magnets, whose magnetization states can be adjusted by d-axis current pulses [1-2]. The PM excitation field is adjustable and the need for continuous flux-weakening current is reduced. To maximize output torque and operation efficiency, the maximum torque per ampere (MTPA) control is widely used in interior PMSMs [3-4]. However, the parameters, like PM flux-linkage, d- and q-axis inductances, of variable-flux PMSMs are influenced by not only the load condition variations but also the magnetization state adjustments. The precise MTPA control for variable-flux PMSMs is highly affected by the nonlinear parameter properties. The operating principle of a novel variable-flux PMSM is analysed. The nonlinear load flux-linkage model is established to simplify the modelling of nonlinear parameter properties. The MTPA current trajectory search method based on Golden Selection method is proposed to obtain the current trajectory considering the influence of magnetization state adjustments and load condition variations. Moreover, the on-line MTPA control method based on neural-network algorithm is proposed to provide an effective way of on-line MTPA control for variable-flux PMSMs. II Operating Principle of Variable-flux PMSMs and Nonlinear Load Flux-Linkage Model A novel series-parallel-connected hybrid-PM variable-flux PMSM is proposed and shown in fig. 1. The magnetization state of this machine can be adjusted by manipulating the magnetization direction of AlNiCo near d-axis in V-shape magnets and the flux amplitude of all AlNiCo. The machine saturation degree varies significantly with magnetization state adjustments and load condition variations. Then the machine parameters, such as PM flux-linkage, d- and q-axis inductances, vary nonlinearly too. The nonlinear load flux-linkage model is established to simplify the modelling of nonlinear parameter properties. The amplitude-phase sequences of current working points are selected to calculate the corresponding d-, q-axis flux-linkage by FEA or experiment. And the permeability of any two adjacent selected points is nearly the same. The nonlinear load flux-linkage model is formed using the selected current working points and corresponding d-, q-axis flux-linkage. The separation of load PM flux-linkage and calculation of d-, q-axis inductances are no longer needed. The load flux-linkage model that presents the nonlinear parameter properties is used in the MTPA current trajectory search process. III MTPA Current Trajectory Search based on Golden Selection Method The proposed MTPA current trajectory search method includes two iterative loops which are both based on Golden Selection method. The current phase iterative loop is implemented firstly, and the optimization target is to minimize corresponding current amplitude at a given torque reference and a given magnetization state. The amplitude is optimized to the minimum value when the iteration interval of phase is within a given minuscule number. However, the corresponding amplitude of phase cannot be obtained directly due to nonlinear parameter properties. The current amplitude iterative loop is nested in the phase iterative loop to obtain corresponding amplitude of each phase. The target of amplitude iterative loop is to optimize the difference between present torque and torque reference to zero. During the amplitude iteration process, the value of present torque under different amplitudes and phases is calculated by the nonlinear load flux-linkage model. MTPA current trajectories of the proposed machine under 4 magnetization states are obtained by the proposed search method, and compared with the trajectories obtained by traditional method, as shown in fig. 2. The current trajectories are evaluated by FEA and experiments. The results prove that the proposed search method can find MTPA current trajectories precisely considering the influence of magnetization state adjustments and load condition variations. IV On-Line MTPA Control based on Neural-Network Algorithm The on-line generation of current commands by the proposed search method is time-consuming. The neural-network-based MTPA control method is proposed to implement MTPA control in real-time control system. The MTPA current trajectory generated by proposed search method under different magnetization states and torque references are selected as sample data. The on-line MTPA control neural-network model is trained, tested and validated using sample data. The model input is magnetization state and torque reference, and the output is current amplitude and phase. Once the train, test and validation process are finished, the knowledge about input-output relationship is acquired and the on-line MTPA control neural-network model is established. MTPA current trajectories at any magnetization state and torque reference can be predicted by this model. The mathematical expressions of established model are easily programmed and quickly executed by DSP for on-line MTPA control. The calculation speed and accuracy are evaluated by simulations and experiments. The results prove that the proposed neural-network model can realize on-line MTPA control for variable-flux PMSMs quickly and precisely.

**References:**

[1] V. Ostovic. "Memory Motors," IEEE Ind. Appl. Mag., 9(1): 52-61, 2003. [2] H. Hua, Z. Q. Zhu, A. Pride, et al. "Comparative Study on Variable Flux Memory Machines with Parallel or Series Hybrid Magnets," IEEE Trans. Ind. Appl., 55(2): 1408-1419, 2019. [3] S. Kim, Y. D. Yoon, S. K. Sul, et al. "Maximum Torque per Ampere (MTPA) Control of an IPM Machine Based on Signal Injection Considering Inductance Saturation," IEEE Trans. Power Electron., 28(1): 488–497, 2013. [4] J. H. Chen, J. Li and R. H. Qu. "Maximum-Torque-per-Ampere and Magnetization-State Control of a Variable-Flux Permanent Magnet Machine," IEEE Trans. Ind. Electron., 65(2): 1158–1169, 2018.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/dfd2667da240b269f350b75ad5c8e9cf.jpg)
Fig.1 Structure of proposed machine.

![](https://assets.underline.io/uploads/markdown_image/1/image/a7ca35c45edd083408f18282dbf1d1bf.jpg)
Fig.2 Current trajectories obtained by proposed method and traditional method.

 Maximum Torque per Ampere Control for Variable-Flux PMSMs Considering the Influence of Magnetization State Adjustments and Load Condition Variations

This digest investigates the flux intensifying (FI) characteristics for permanent magnet assisted synchronous reluctance motor (PMa-SynRM) to achieve high torque density with a small amount of PM. This motor is particularly denoted as "FI-PMa-SynRM". The developed FI-PMa-SynRM is then compared with two other counterparts, i.e., a surface permanent magnet synchronous motor (SPMSM) and a synchronous reluctance motor (SynRM). The analysis is conducted using finite element method through flux linkage and inductances so that the effectiveness of FI characteristics is demonstrated. The analysis shows that a high torque density (24.8 Nm/L) can be achieved with a small amount of PM (0.72% motor volume). I. Motor Models Flux intensifying (FI) rotor configuration has been of recent interest as an alternative to conventional permanent magnet synchronous motors (PMSMs) which usually adopt flux weakening feature [1], [2]. By using a moderate amount of PM, the FI configuration can be applied to synchronous reluctance motor (SynRM), which leads to the introduction of FI permanent magnet assisted SynRM (FI-PMa-SynRM) [3]. In order to minimize the amount of PM used, a surface-inset PM design is employed, indicating that the rotor configuration can be considered as the combination of a SynRM with multiple flux barriers (FBs) and a surface PMSM (SPMSM), as shown in Fig. 1(a). This configuration (PM placed along d-axis) differs from common PMa-SynRM (PM placed along q-axis) so that PM flux linkage is intensified by part of the armature flux. This is called "FI" as illustrated in Fig. 1(b), making PM not easily demagnetized in such a configuration. Two other models with the same rotor shapes are used for comparison: the first one is an inset SPMSM (only eliminating interior FBs when compared to the developed FI-PMa-SynRM) that represents the flux weakened motor while the second one is a SynRM (only removing PMs based on the FI-PMa-SynRM) with only reluctance torque, as indicated in Fig. 1(a). The motor specifications are presented in Table I. II. Analysis and Results A. Distribution of magnetic field - The no-load condition is considered first, where the flux lines and flux densities are shown in Fig 1(c). It can be seen that for the FI-PMa-SynRM the flux focuses on the region between two adjacent interior FBs. For the inset SPMSM the flux focuses around the PMs. However, the PM operating conditions are similar since the flux density distributions in the PMs are basically affected by their relative positions to the stator slots. - Fig. 1(d) shows the motor flux densities under the maximum torque per ampere (MTPA) control where the FI-PMa-SynRM and SynRM operated with negative current phase advance angles with the SynRM having a greater phase advance. These all correspond to the phasor diagrams presented in Fig. 1(b). - The PM flux densities of the FI-PMa-SynRM and inset PMSM at -25 degrees (current phase advance) based on MTPA control of FI-PMa-SynRM are shown in Fig. 1(e). The negative current phase advances indicate that these motors operate in the FI mode. As can be seen, the overall flux densities in the PM seem to increase with applied current although part of the PMs may still suffer lower operating points. However, when the inset SPMSM operates in the MTPA condition, a positive current phase advance (19 degrees) is required, and the PM operating points are much lower due to flux weakening, which is also shown in Fig. 1(e). These imply that the PM is more secure with FI and the torque density can be possibly improved without suffering demagnetization. B.Inductance variation The inductances vs current amplitude when applying current on the d-axis are shown in Fig. 1(f). It can be seen that, for both the FI-PMa-SynRM and SynRM, Ld is higher than Lq so that the saliency, i.e. Ld/Lq of these models is higher than 1. Nevertheless, Ld of the SynRM is rapidly reduced with current increase while that of the FI-PMa-SynRM maintains. For the inset SPMSM, Ld is slightly lower than Lq and they both decrease when current magnitude increase so that the saliency is slightly lower than 1. C.Torque production Fig. 1(g) shows the torque vs current angle of these motors at both rated and peak currents. The MTPAs of the FI-PMa-SynRM and SynRM are achieved with negative current phase advance and that for the inset SPMSM is achieved with a positive phase advance. Moreover, with an added PM amount of just 0.72% of motor volume, the FI-PMa-SynRM has 1.77 and 1.34 times torque of that of the SynRM at rated and peak currents, respectively. On the other hand, by adding some FBs, the FI-PMa-SynRM produces 1.4 and 1.73 times torque of that of the inset SPMSM at the rated and peak currents, respectively. This indicates that the FI-PMa-SynRM is capable of effectively combining reluctance torque or PM torque compared to the other types of motors. As a result, this motor achieved high torque density, i.e. 13.45 Nm/L and 24.8 Nm/L at rated and peak conditions, respectively. D. Measurements The experimental results based on the prototype shown in Fig. 1(a) will be reported in the full paper. III. Conclusion This digest compares the characteristics of the designed FI-PMa-SynRM and its counterparts. The results revealed that the target FI-PMa-SynRM can achieve higher torque density than its counterparts by using a small amount of PM.

**References:**

[1] N. Limsuwan et al, IEEE Trans. Ind. Appl., vol. 50, no. 2, pp. 1015–1024, Mar. 2014. [2] M.-T. Chui et al, 2015 18th International Conference on Electrical Machines and Systems (ICEMS), Oct. 2015, pp. 232–236. [3] D.-K. Ngo, M.-F. Hsieh, and T. A. Huynh, IEEE Trans. Magn., vol. 55, no. 7, pp. 1–8, Jul. 2019.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/2d6c81d432c420ef48d64af683fd1974.jpg)

![](https://assets.underline.io/uploads/markdown_image/1/image/7b37cb7115558e9622ed8f166329d82d.jpg)

 Flux Intensifying Features of Permanent Magnet Assisted Synchronous Reluctance Motor with High Torque Density

I. INTRODUCTION The traditional three-phase PMSMs are confronted with severe challenge, and some researchers turn to investigate modular motor systems for better fault-tolerant capability [1]. The modular motor system is composed of several modules, and each module could be equivalent to an independent sub-motor. The better the independence of electric, magnetic, thermal, and physics between the modules, the better fault-tolerant performance of the motor is. The fractional slot concentrated winding (FSCW) has some advantages in non-overlapping end winding, large inductance and small coupling between phases, so it is very suitable for the application in modular fault-tolerant motors [2] However, the MMF of the motor with FSCW contains quantities of space harmonics which will cause PM eddy-current losses. The open-circuit fault is a typical fault in all kinds of motor faults [3], and the fault-tolerant capability of the motor is mostly determined by the output performance under open-circuit faults [4]. This paper investigates the influence of magneto motive force (MMF) space harmonic on permanent magnet (PM) eddy-current losses in a modular fault-tolerant in-wheel motor under various open-circuit faulty operations. First, reconfiguring maximum round MMF with constant amplitude of current is taken as the fault-tolerant (FT) strategy for the open-circuit of one phase. The fault-tolerant strategies for various faults are derived respectively based on this strategy. Second, the harmonic components of MMF under various faults and corresponding fault-tolerant control are analyzed respectively. Then, an analytical model based on multi-loop method for PM eddy current losses is proposed. The losses induced by each harmonic components of MMF can be obtained from the model. The analytical results agree well with the finite element method (FEM) results. The space harmonics are composed of slot harmonics and other harmonics. The proportions of eddy current losses caused by these two types of harmonics under various faults are derived. Finally, PM eddy-current losses induced by the fault-tolerant current under the overload condition are analyzed, and the results have a certain reference value for the fault-tolerant strategy which makes the faulty motor operates safer and more stable. II. MODULAR MOTOR SYSTEM WITH FAULT-TOLERANT CAPABILITY For design a better fault-tolerant motor, the entire fault-tolerant system needs to be considered comprehensively. Not only design the motor itself, but also should consider the topology of the controller and the fault-tolerant control strategy. The motor consists of 4 modules, as shown in Fig. 1. Each module is controlled by an independent drive with a three-phase four-wire control circuit, which ensures electrical isolation between the modules. III. ANSLYSIS OF CHARACTERISTICS OF MMF UNDER VARIOUS OPEN-CIRCUIT FAULTY OPERATIONS 1) Reconfiguring Maximum Round Fundamental MMF With Constant Current When A1 is in open-circuit condition, the performance of MMF will be declined. Quantities of sub-harmonics and high order harmonics closed to the working wave emerge, and the amplitude of working wave decreases, as shown in Fig. 2. Through the proposed fault-tolerant control, the performance of MMF is improved. 2) Spatial Harmonic Analysis of MMF under Various Faults The overall performance evaluation of the MMF is established by the two following aspects: 1 The amplitude of the working wave; 2 Total harmonic distortion (THD). The performance of MMF for the modular motor under various open-circuit faults can be effectively improved through the proposed fault-tolerant strategies. IV. ANSLYSIS OF PM EDDY-CURRENT LOSSES UNDER VARIOUS LOSSES UNDER VARIOUS OPEN-CIRCUIT FAULTY OPERATIONS 1) Analytical Model An analytical model based on multi-loop method for PM eddy current losses is proposed. 2) Verification by FEM The eddy-current losses of PMs by FEM consist of two parts. One is the slot ripple losses in PMs, which are the most important losses in PMs and generated by the fundamental frequency magnetic flux due to the stator slot openings. The other one is the losses in PMs induced by the space harmonics. The analytical results agree well with the FEM results. 3) Analysis of PM Eddy-Current Losses under Overload Fault-Tolerant Control When the overload FT controls are adopted, the motor can operate safely for a period of time rather than a long time. If we want the motor operates stably during the time, the PM losses should be considered. V. CONCLUSIONS An analytical model based on multi-loop method for calculating PM eddy-current losses is proposed. The analytical results agree well with the FEM result, which proves the correctness of the model. The proposed model is more accordant with the actual situation and the physical meaning is clearer. The eddy-current losses caused by high order slot harmonics are very small. Among the newly emerged harmonics, the sub-harmonics and high order harmonics closed to working wave will cause relatively large eddy-current losses. As the fault level increases, the proportion of eddy-current loss induced by slot harmonics decreases, while the proportion of eddy-current loss induced by other harmonics increases. If the overload fault-tolerant strategy is adopted, not only the copper losses but also the eddy-current losses of PMs should be considered.

**References:**

[1] Ifedi C. J., Mecrow B. C., Brockway S. T. M, “Fault-tolerant in-wheel motor topologies for high-performance electric vehicles”. IEEE Trans. Ind. Appl., 2013, 49, (3), pp. 1249-1257. [2] Heins G., Ionel D., Thiele M., “Winding factors and magnetic fields in permanent magnet brushless machines with concentrated windings and modular stator cores”. IEEE Trans. Ind. Appl., 2015, 51, (4), pp. 2924–2932. [3] Li Y. X., Zhu Z. Q., Thomas A. S., “Novel modular fractional slot permanent magnet machines with redundant teeth”. IEEE Trans. Magn., 2019, 55, (9), pp. 1-10. [4] Liang P. X., Pei Y. L., Chai F., “An improved method for armature-reaction magnetic field calculation of interior permanent magnet motors”. IEEE Trans. Magn., 2016, 52, (7), pp. 1-4.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/b5f8d38fe4176b0605a175e03295d3aa.jpg)

![](https://assets.underline.io/uploads/markdown_image/1/image/3dbccfaa57183c0d06276f162998ca2a.jpg)

Next from INTERMAG 2021

Spin-orbit torque rectifier for energy harvesting from weak radio-frequency

Similar lecture

Design and Analysis of a V Shaped Leakage