Breakthrough in EV Motor Design: Multi-Layer Magnet IPMSM Achieves High Efficiency and Low Noise
In the rapidly evolving landscape of electric mobility, where performance, efficiency, and driving comfort are paramount, a new research study has introduced a significant advancement in permanent magnet synchronous motor (PMSM) technology. A team of researchers from the School of Electrical and Information Engineering at Anhui University of Science and Technology has developed and optimized a novel multi-layer built-in permanent magnet synchronous motor specifically designed for hairpin winding applications in electric vehicles (EVs). This innovative motor design promises to deliver higher power density, improved efficiency, and notably reduced torque ripple and cogging torque—three critical factors that directly influence the driving experience and overall performance of modern EVs.
The research, led by Dr. Han Lin, Associate Professor and Master’s Supervisor, along with graduate student YuHang Zhang, ZhongGen Wang, and TianLong Deng, has been published in the Journal of Chongqing University of Technology (Natural Science). Their work presents a comprehensive approach to motor design that integrates advanced electromagnetic modeling, finite element analysis, and sophisticated optimization techniques to push the boundaries of what is possible in compact, high-performance electric drive systems.
As the global automotive industry continues its shift toward electrification, the demand for more efficient and powerful electric motors has never been greater. Traditional motors, while reliable, often face limitations in terms of power density, thermal management, and acoustic performance. Hairpin winding technology has emerged as a promising solution to some of these challenges, offering higher slot fill factors, reduced AC losses, and improved heat dissipation compared to conventional round-wire windings. However, even with the advantages of hairpin winding, the design of the rotor and its magnetic structure remains a critical area for innovation.
The research team focused on the rotor architecture, specifically exploring multi-layer magnet configurations within an interior permanent magnet (IPM) layout. Unlike conventional single- or double-layer designs, the proposed motor features a three-layer magnet structure that combines U-shaped, V-shaped, and I-shaped permanent magnets in a unique UV-I configuration. This complex arrangement allows for greater control over the magnetic flux distribution within the motor, enabling engineers to fine-tune performance characteristics such as torque production, flux weakening capability, and harmonic content.
One of the primary challenges in high-speed electric motors is the presence of torque ripple and cogging torque, which can lead to vibrations, noise, and reduced ride comfort. These undesirable effects are particularly pronounced in interior permanent magnet motors due to the interaction between the stator teeth and the rotor’s magnetic poles. To address this, the researchers implemented a multi-stage optimization strategy that begins with rotor segmentation and skewing.
Segmented skewing is a well-known technique used to reduce electromagnetic harmonics by introducing a controlled misalignment between rotor segments. In this study, the team adopted a linear segmented skew design with three segments and a skew angle of 5 degrees. This initial optimization step yielded remarkable results: cogging torque was reduced by an impressive 97.01%, while torque ripple dropped from 9.91% to 3.18%. Although there was a slight increase in the total harmonic distortion (THD) of the air gap flux density, the overall improvement in smoothness and noise reduction justified the trade-off.
With the baseline performance significantly enhanced through skewing, the researchers proceeded to compare four different multi-layer magnet configurations: UU-I, UU, UV, and UV-I. Each design was evaluated based on key performance indicators such as average torque, torque ripple, back electromotive force (EMF) THD, and air gap magnetic flux density. Among the four, the UV-I configuration stood out as the most balanced and effective. It delivered an average torque of 129.19 N·m with a torque ripple of just 3.18%, while maintaining a relatively low cogging torque of 244.11 mN·m. The back EMF waveform exhibited excellent sinusoidal characteristics with a THD of only 2.88%, indicating minimal harmonic distortion and smoother operation.
Having identified the UV-I structure as the optimal candidate, the next phase of the research focused on fine-tuning the dimensions of the individual magnet segments. This is where the team employed the Taguchi method, a powerful statistical optimization technique widely used in engineering design to identify the most influential parameters and determine their optimal settings with a minimal number of experiments.
Six key geometric parameters were selected for optimization: the width and total length of the U-shaped magnet, the width and total length of the V-shaped magnet, and the width and total length of the I-shaped magnet. Each parameter was assigned five different levels, resulting in a total of 25 experimental combinations based on an orthogonal array. Finite element simulations were conducted for each combination to evaluate average torque and torque ripple.
The results of the Taguchi analysis revealed that the length of the V-shaped magnet (D) had the most significant impact on average torque, contributing 56.65% to the overall variation. This was followed by the length of the U-shaped magnet (B) at 20.45% and the width of the I-shaped magnet (E) at 14.53%. For torque ripple, the width of the I-shaped magnet (E) emerged as the most influential factor, accounting for 43.13% of the variation, followed by the total length of the I-shaped magnet (F) at 31.48%.
Based on these insights, the researchers selected the optimal combination of parameter levels: A(5), B(5), C(4), D(4), E(3), F(5). This configuration corresponds to specific dimensional values that maximize average torque while minimizing torque ripple. After applying this optimized geometry, the motor’s performance improved further: average torque increased to 131.28 N·m, torque ripple dropped below 2% to 1.78%, and the fundamental amplitude of the back EMF rose from 274.09 V to 279.35 V. However, an unexpected side effect was observed—the cogging torque increased slightly from 244.11 mN·m to 259.78 mN·m. This indicated that while the geometric optimization successfully enhanced torque output and smoothness, it introduced a new challenge in terms of magnetic detent forces.
To address this issue, the team turned to an advanced magnetization technique known as Halbach array magnetization. A Halbach array is a special arrangement of permanent magnets that concentrates the magnetic field on one side while canceling it on the other. In the context of electric motors, partial Halbach magnetization can be used to shape the air gap flux distribution, reducing harmonics and minimizing cogging torque without sacrificing torque density.
The researchers applied Halbach magnetization selectively to different layers of the multi-layer rotor. They tested four configurations: magnetizing the I-shaped magnet, the V-shaped magnet, the inclined segment of the U-shaped magnet, and the horizontal segment of the U-shaped magnet. Each was evaluated under a fixed initial magnetization angle of 35 degrees, with the length of the magnetized segment varied parametrically to find the optimal setting.
The results were revealing. While all configurations showed some degree of cogging torque reduction, the V-shaped magnet demonstrated the most consistent and effective performance. As the length of the Halbach-magnetized segment increased, the suppression of cogging torque improved steadily. More importantly, unlike the U-shaped magnet’s inclined segment—which caused a significant rise in torque ripple—the V-shaped magnet maintained a low torque ripple throughout the optimization process.
Encouraged by these findings, the team focused exclusively on optimizing the Halbach magnetization of the V-shaped magnet. They conducted a second round of parametric scanning, this time varying both the magnetization angle (from 25° to 45°) and the magnetized segment length (from 4 mm to 16 mm). The goal was to find a balance between maximum cogging torque reduction and minimal impact on torque ripple and average torque.
The analysis showed a clear trend: as the magnetization angle increased, so did the effectiveness of cogging torque suppression. However, this came at a cost—torque ripple also increased, and average torque began to decline. At a magnetization angle of 45° and a segment length of 9 mm, cogging torque reached its lowest point, reduced by 45.08% compared to the post-Taguchi optimization state. However, torque ripple rose above 2%, and average torque dropped by 4.59%.
Seeking a more balanced solution, the researchers identified a sweet spot at a magnetization angle of 30° and a segment length of 9 mm. At this setting, cogging torque was reduced by 25.43%, torque ripple remained exceptionally low at 1.72%, and average torque decreased by only 2.26% to 128.31 N·m. This configuration represented the ideal compromise, delivering substantial noise and vibration reduction without compromising the motor’s core performance metrics.
The final optimized motor demonstrated impressive overall characteristics. Under full load with a current of 400 A, the efficiency map revealed a peak efficiency of 96.74%, with a broad high-efficiency zone covering a wide range of operating conditions. The motor achieved a rated power of 134.35 kW at 10,000 rpm and a peak power of 298 kW, with a maximum speed of 22,000 rpm. These figures highlight the motor’s capability to deliver both sustained performance and high dynamic response—essential qualities for modern EVs.
To validate the superiority of their design, the researchers compared it with a previously published hairpin-wound double-layer V-shaped IPMSM under similar specifications (8 poles, 48 slots, comparable physical dimensions). The results were striking: the newly developed multi-layer UV-I motor outperformed the reference design in nearly every category. It achieved a higher rated torque (128.31 N·m vs. 82.9 N·m), greater peak power, higher maximum speed (22,000 rpm vs. 12,000 rpm), and superior efficiency. This comparative analysis underscores the tangible benefits of the multi-layer magnet architecture combined with advanced optimization techniques.
From a manufacturing and practical standpoint, the proposed design maintains compatibility with existing hairpin winding production processes, ensuring scalability and cost-effectiveness. The use of segmented skewing and selective Halbach magnetization does introduce additional complexity, but the researchers argue that the performance gains justify the added engineering effort, especially for premium EV applications where refinement and efficiency are key selling points.
Moreover, the study contributes valuable insights into the role of individual magnet shapes within multi-layer rotors. It confirms that V-shaped magnets offer the most balanced influence on motor performance, enhancing both torque production and flux controllability. The I-shaped magnet, while reducing torque and flux levels, plays a crucial role in suppressing harmonics and improving waveform quality. The U-shaped magnet, though powerful in boosting torque, tends to increase ripple and distortion, requiring careful tuning to avoid negative side effects.
This research exemplifies the kind of multidisciplinary engineering required to advance electric propulsion technology. By combining electromagnetic theory, computational modeling, statistical optimization, and advanced magnetization strategies, the team has created a motor that not only meets but exceeds the demanding requirements of next-generation electric vehicles. The integration of hairpin winding with a meticulously optimized multi-layer rotor represents a significant step forward in the pursuit of higher efficiency, lower noise, and greater power density.
As automakers continue to push the limits of EV performance, innovations like this will play a crucial role in shaping the future of sustainable transportation. The work conducted by Han Lin, YuHang Zhang, ZhongGen Wang, and TianLong Deng at Anhui University of Science and Technology demonstrates that even within the well-established framework of permanent magnet motors, there remains ample room for breakthrough improvements through thoughtful design and rigorous optimization.
Published in the Journal of Chongqing University of Technology (Natural Science), Vol. 38, No. 7, 2024. DOI: 10.3969/j.issn.1674-8425(z).2024.07.029