Abstract
The optimized design of a new high-temperature superconducting rotating pole machine is presented. Its main structural feature is the use of a double stator core which separates the synchronous machine pole shoe from the pole body to rotate separately as the machine rotor, allowing the superconducting coil to operate in a stationary state. The inner stator core, the stationary dewar and the rotor core together form the excitation system of the machine. The excitation coil windings adopt a rectangular cross-section, with flux divertor strategically placed between the high-temperature superconducting coils. This configuration aims to modulate the background magnetic field, specifically reducing the perpendicular magnetic field component. This mitigation minimizes the impact of ambient magnetic fields on the superconducting coil’s current carrying capacity, ensuring an optimized magnetic field environment for its operation. Through the integration of these modifications, the technical and economic parameters of the enhanced high-temperature superconducting machine have been significantly improved. The optimization of design, coupled with detailed calculations of the 3D electromagnetic field, was achieved utilizing the commercial software Ansys EM module.
Keywords
Introduction
In recent years, wind power generation has become the leading method of producing new energy due to the growing concerns over environmental pollution and energy depletion. As wind power generation continues to expand, the demand for high power density and highly efficient machines has become increasingly urgent [1]. Among them, the high-temperature superconducting machine (HTSM) utilizes superconducting strips with no resistance and excellent current-carrying capabilities as the machine windings. This drastically boosts the machine’s power density, lowers its weight, and enhances its operational efficiency [2,3]. Meanwhile, the ongoing R&D and production of second-generation high-temperature superconducting strips (YBCO) have led to crucial advancements in critical current, strip length, and mechanical structure performance. This has significantly boosted the competitiveness of high-temperature superconducting machines, attracting numerous research institutions to delve into the subject [4].
Due to the limitations of high-temperature superconducting strip temperature, magnetic field, and critical current, and to ensure the efficiency of HTSM, most existing high-temperature superconducting machines are designed as “semi-superconducting” machines using superconducting coils for excitation and conventional copper wires for armature windings. For example, the 2 MW superconducting direct-drive wind turbine of the 712th Research Institute of China Shipbuilding Group [5], the 3 MW ship propulsion machine [6], the 10 MW wind turbine of Japan [7], and the 500 kW generator of the Institute of Electrical Engineering of the Chinese Academy of Sciences [8] are all in the form of semi-HTSM structure. Although some megawatt-scale superconducting generator prototypes have been successfully developed based on this structure, the reliability is difficult to be guaranteed due to the use of complex cooled cryogenic rotating dewar. In addition, the transmission of current and monitoring signals relies on rotating elements such as brush slip rings, resulting in high maintenance costs and low reliability. On this basis, a High Temperature Superconducting Machine with a Rotating Pole-shoe (RPHTSM) with YBCO strips is proposed as a research object to achieve static sealing of coolant while avoiding the use of complex structures such as brushes and slip rings, and improving the reliability of current and monitoring signals. At the same time, it avoids the use of complicated structures such as brushes and slip rings, and improves the reliability of the superconducting machine structure. In order to improve the current carrying capacity of the superconducting field coil, it is necessary to optimize the magnetic leakage distribution within the machine. This in turn reduces the effect of the perpendicular magnetic field of the superconducting strip on the current carrying capacity, and a common method is to add a flux divertor [9–11].
This study focuses on the excitation coil system of a High-Temperature Superconducting Machine with a Rotating Pole-shoe (RPHTSM), utilizing YBCO strips. A comprehensive three-dimensional magnetic field analysis model for the RPHTSM is developed. The investigation delves into the intricate impact of magnetic leakage on the superconducting coil’s performance through meticulous three-dimensional finite element calculations. Special attention is given to the influence of perpendicular magnetic fields within the background field on the superconducting coil’s efficacy.
To optimize the superconducting rotating pole-shoe machine, an innovative approach integrates a flux divertor within the superconducting coil. This design variation is explored to analyze its effect on the background magnetic field and, consequently, on the critical current decay of the superconducting strip. The goal is to enhance machine performance and refine its structure.
Results demonstrate that the inclusion of a flux divertor significantly diminishes the perpendicular magnetic field component within the superconducting strip. Furthermore, three-dimensional static magnetic field simulations explore various flux divertor structure designs, specifically evaluating the required superconducting strip material under different configurations. Notably, the simulations validate that the optimized high-temperature superconducting machine exhibits considerable enhancements in techno-economic parameters.
High-temperature superconducting strips and critical currents
High-temperature superconducting materials, since their discovery, have found wide-ranging applications in various superconducting power equipment, spanning from superconducting cables to energy storage, transformers, and current limiters. These materials boast several advantages, including high critical current density, minimal AC losses, and the capacity to operate within the liquid nitrogen temperature range. The second-generation YBCO high-temperature superconducting strips offer superior current-carrying capabilities compared to their first-generation counterparts at 77 K, marking a significant recent development. Notably, these second-generation strips play a pivotal role in downsizing and enhancing the performance of machines.
Critical current stands as a crucial performance metric for superconducting machines, where in these machines generate their magnetic field during operation. The critical current of high-temperature superconducting strips diminishes under the influence of the background magnetic field, with the extent of decay determined by the direction and intensity of this field. The magnitude of critical current significantly impacts the superconducting machine’s performance, emphasizing the importance of investigating the actual decay of critical current within the high-temperature superconducting strip under the influence of the background magnetic field. Given the anisotropic physical structure inherent in superconducting strips, variations in electromagnetic properties exist across different directions.
This study initiates by selecting the ST05EL/100 strip, manufactured by Shanghai Superconductor Technology, as the primary material for designing the high-temperature superconducting machine with rotating pole shoes. Table 1 presents the specific parameters of this chosen strip material as provided by the manufacturer.
Superconducting strip parameters
Superconducting strip parameters
The superconducting strip exhibits remarkable critical current density, with its peak performance reaching 600 A/cm at 77 K. Figure 1 depicts the normalized critical current as a function of perpendicular and parallel magnetic fields for a 77 K superconducting strip utilizing liquid nitrogen as the cooling medium. This data is derived from the ST05EL/100 superconducting strip supplied by Shanghai Superconductivity Company. The normalized critical current value is defined as per [12]:

Critical current characteristic curve of ST-05-EL100 [12].
As can be seen from Fig. 1, the attenuation of the critical current of superconducting strips gradually increases with the gradual increase of the perpendicular magnetic field, and the attenuation of the critical current is obvious when the perpendicular magnetic field increases from 0 to 0.5 T. From 0.6 T onwards, the influence of the magnetic field on the current gradually decreases to only about 25% of the initial value of the self-field. This shows that the critical current is highly dependent on the magnetic field strength. In other words, the strip critical current is highly susceptible to the influence of the magnetic field perpendicular to the strip surface.
The RPHTSM is a 3-phase, 4-pole, 690 V, 500 kW synchronous generator with the structure shown in Fig. 2, which mainly includes the housing, outer stator core, armature winding, rotor, dewar assembly, superconducting coil and inner stator core.

Structure of RPHTSM.
The RPHTSM, an advancement upon conventional synchronous machines, introduces a unique design with a second air gap, separating the synchronous machine’s pole shoe from its pole body to function as the machine rotor. This innovation allows the superconducting coil to operate while stationary, creating the machine’s excitation system in collaboration with the inner stator core, the stationary dewar, and the rotor core. Distinguishing itself through a modified magnetic circuit structure, the RPHTSM’s configuration involves the separate rotation of the pole shoe as the rotor, distinct from the magnetic pole core. Through structural adjustments, the pole shoe and superconducting coil transform into an inner stator, comprising a superconducting coil and an iron core. This inner stator, in conjunction with the rotor, forms an excitation circuit facilitated by the second air gap, collectively constituting the machine’s excitation system.
The utilized excitation system ensures the stationary operation of the superconducting coil, thereby reducing the rotor’s inertia. The primary magnetic flux, originating from the coil’s N-pole, traverses through the inner stator core, the second air gap, the rotor’s N-pole, and the first air gap before reaching the outer stator. Its cycle completes by returning to the coil’s S-pole through the external stator, the first air gap, the rotor, the second air gap, and the internal stator. The machine’s magnetic field is characterized by axial and radial components. The rotor core, composed of alternating N and S poles connected by non-conductive structural elements, generates this magnetic field. This field interacts with the armature’s rotating field, facilitating electromechanical energy conversion. Liquid nitrogen and power are supplied via a hollow shaft from the machine end, eliminating the need for brushes and rotating dynamic seals.
The machine’s structure creates a intricate internal leakage field characterized by a well-defined three-dimensional distribution. This field significantly influences the background magnetic field within the superconducting coil during operation, consequently impacting the current-carrying capacity of the superconducting strip. It is imperative to thoroughly investigate the machine’s leakage field, elucidate the excitation current of the superconducting coil, and optimize the coil’s structure to mitigate the impact of magnetic leakage on the carrying capacity of the superconducting strip.
Enhancing the Current-Carrying Capacity of Superconducting Machine Excitation Systems is Paramount. The operational critical current of high-temperature superconducting strips is significantly constrained by their inherent anisotropy, posing economic challenges. Consequently, determining the excitation current for the superconducting coil holds exceptional importance in the design of a superconducting machine’s excitation system.
Figure 1 illustrates the experimental test outcomes displaying the critical current at a temperature of 77 K under varying magnetic fields. In this representation, 0° and 90° respectively denote the critical current values of the superconducting strip when subjected to magnetic fields parallel and perpendicular to the strip plane. For a visual depiction of the perpendicular and parallel magnetic fields at the surface of the superconducting strip, refer to Fig. 3.

Superconducting strip cross-sectional perpendicular and parallel magnetic field distribution.
The critical current calculation of the high-temperature superconducting excitation winding based on the domestically produced YBCO strip is firstly based on the Kim-like model shown in Eq. (2) to obtain the approximate critical current under the magnetic field [13].
When |B| ≤ 0.003 T,
When |B| > 0.003 T,
The critical current of high-temperature superconducting wire exhibits a gradual decrease as the magnetic field intensifies. Despite facing high magnetic fields, the wire still conducts a substantial current. However, due to the inherent instability in the mechanical structure of high-temperature superconducting strips, utilizing the conventional solenoidal layer-winding method leads to significant performance decay.
To address this, the excitation coil, positioned between the inner stator and rotor, is wound using a disc coil configuration. This alteration ensures that the critical current of the high-temperature superconducting wire within the disc coil is contingent upon the perpendicular magnetic field applied across its wide surface.
In constructing excitation coils via stacked disc coils, each disc coil’s critical current varies due to differing perpendicular magnetic field intensities. Despite these discrepancies, conventionally connecting all disc coils in series and powering them with a single source limits the current across the coils to the minimum allowable current. Consequently, areas with maximal perpendicular magnetic field distribution, typically situated at the top and bottom disc coils, regulate the maximum current passing through the windings. Thus, the critical current of the excitation coil hinges on the maximum current permitted at the point of peak perpendicular magnetic field within the winding.
Proposal for preliminary conceptual design
The RPHTSM 3D leakage magnetic field calculation model is shown in Fig. 4. The structures related to the magnetic field distribution, such as the outer stator, the inner stator, the superconducting coil excitation system, and the rotor, are mainly established.

Calculation model of RPHTSM.
In order to reduce the computational workload, a 1/4 model for RPHTSM 3D magnetic field analysis was further developed based on ANSYS Maxwell, and the preliminary calculated magnetic field distribution characteristics of the machine are shown in Fig. 5. When current is applied to the superconducting coil, a magnetic field is generated within the machine’s inner core and transmitted along it. This magnetic field enters the rotor core through the second air gap and subsequently extends into the outer stator core through the first air gap. This process induces a voltage in the armature winding.

3D magnetic field distribution of RPHTSM.
In this excitation system structure, the leakage magnetic field in the main magnetic flux of the superconducting coil winding is mainly divided into two main parts: one part of the flux loop is formed between the rotor poles through the air gap as 𝛷1; and one part of the flux loop is formed between the rotor poles and the inner stator as 𝛷2, as shown in Figs 6 and 7. Both parts of the magnetic leakage will indirectly affect the current-carrying capacity of the superconducting strip cross-section. Theoretically, after the use of the magnetic ring structure, the leakage flux through the superconducting coil is reduced due to the magnetization of the magnetic shielding (silicon steel sheet), which makes the leakage flux through the superconducting strip lower, thus reducing the stray loss.

Inter-pole magnetic leakage of the rotor.

Magnetic leakage between rotor and inner stator.
The RPHTSM comprises five high-temperature superconducting field windings interconnected in series for ease of installation. To ensure overall convenience in installation, each superconducting excitation winding maintains an equal number of turns. The superconducting strip utilized in each branch is the ST05EL/100 strip manufactured by Shanghai Superconductor, boasting a nominal critical current of >200 A at 77 K under self-field conditions, with dimensions of 4.75 mm width and 0.26 mm thickness. The manufacturing process for the disc-shaped coils involves first producing the magnetic shunt spacer, followed by reshaping the flat HTS coil by pressing it into the magnetic shunt spacer and bonding it with resin. In the preliminary machine design, an excitation system without embedded flux divertors is employed, and each superconducting coil is wound around a G10 glass fiber skeleton to enhance its mechanical strength.The size of the coils depends on the volume of the space between the rotor and the inner stator. The radial space between the coil and the Dewar structure will be in the range of 40–55 mm, while the axial space will be in the range of 200 mm.
The steps for designing the excitation coil are as follows: initiate the process by establishing the excitation electromotive force for the machine. Taking into account the machine dimensions and initial values, perform calculations to assess machine performance and determine the no-load induced electromotive force. Through iterative calculations, the final machine design sets the total magnetization ampere-turns at 20,000 ampere-turns, ensuring that the no-load induction potential falls within the allowable error range and the rotor core is in a state of saturation. The initial design involved applying a current of 100 A per superconducting strip, with a total of 40 turns in each superconducting winding. Each high-temperature superconducting excitation winding carried 4000 A. Consequently, the total length of the required HTS strip for this prototype was approximately 199.70 m.
The simulation accuracy is improved, and the simulation time is reduced by dissection of the solution domain into tetrahedrons and encryption of the superconducting coil region. Figure 8 demonstrates that the mesh dissection quality in the region of the superconducting coil can reach 1mm, satisfying the quality standards. Figures 8(a) and 8(b) depict the comprehensive distribution of perpendicular magnetic field components using the complete superconducting strip and 5 coil windings, respectively. When keeping equal amounts of both superconducting strips under control, it is evident that the perpendicular magnetic field component is diminished in the superconducting excitation coil with multiple windings. Consequently, critical current experiences growth with an increase in the number of disc coils.
The five high-temperature superconducting windings are numbered from 1 (top) to 5 (bottom). Simulation results distinctly reveal that the perpendicular magnetic field strength in superconducting excitation coils No. 1 and No. 5 exceeds that at other locations. Furthermore, the calculated actual critical current decay for each superconducting winding is as follows: 31.21 A, 38.68 A, 46.21 A, 39.24 A, and 32.27 A, respectively. The critical current range of the disc coil falls between 31.21 A and 46.21 A. Considering the series connection of various superconducting excitation windings, the current design imposes a maximum allowable critical current of 32.27 A. Complying with this limit necessitates 624.06 m of Superconducting strips for the machine design. Hence, crucial tasks in the optimization design of the RPHTSM include optimizing the distribution of machine magnetic leakage and designing the superconducting excitation structure to mitigate the impact of machine magnetic leakage on the critical current of the superconducting strip.

Superconducting coil sectioning.
As previously discussed, the critical current of the superconducting coil in the RPHTSM excitation system has significantly deteriorated. To limit the critical current value of the strip to an acceptable level, it is crucial to optimally adjust the background magnetic field distribution of the excitation system in order to decrease the magnitude of the perpendicular magnetic field component. Previous research has demonstrated that the implementation of flux divertors in a “semi-superconducting” machine can enhance the distribution of the machine magnetic leakage and decrease the perpendicular magnetic field component in the superconducting coil [15–17]. Consequently, this study incorporates flux divertors into the RPHTSM design to evaluate the magnetic field distribution in the machine and the influence on the current carrying capability of the superconducting coil.
The procedure for designing the RPHTSM excitation system, taking into account flux divertors, involves the following steps: Target Definition: Determine the optimal design target for the high-temperature superconducting excitation coil. This involves limiting the perpendicular magnetic field component of the superconducting strip to a specific range. The lower this value, the higher the design target for the flux divertor structure. Basic Conditions Definition: Specify the fundamental conditions for the optimal design of the flux divertor. This includes selecting superconducting materials, determining the working temperature and cooling mode of the superconducting strip, deciding on the length of the superconducting strip, and selecting the material for the flux divertor. Structural Design: Design various structural dimensions of the flux divertor, create an electromagnetic simulation model, and conduct electromagnetic analysis using optimization algorithms. Validation and Optimization: Check if the perpendicular magnetic field component of the superconducting excitation coil is within acceptable limits after implementing the flux divertor design. If a significant reduction is achieved, conclude the process. If the simulation results do not meet the design goal, modify the geometric parameters of the magnet, and initiate the next round of optimization design.
The optimized design aims to limit the perpendicular magnetic field (normal vector field) in the cross-section of the superconducting strip to an acceptable level while minimizing the total length of the HTS strip. The number of turns and dimensions of the superconducting coil are designed quantitatively, while the shape and dimensions of the flux divertor are treated as variables.
The flux diverters are designed in three different structures, as illustrated in Fig. 9. The presence of various flux diverter structures can influence the perpendicular magnetic field components of the superconducting strip.

Scaled-up design of three flux divertor structures.
The excitation coil of the superconductor is designed to have 2000 turns for each of the flux divertor designs. In order to obtain the perpendicular magnetic field component of each turn of the coil winding, a three-dimensional static magnetic field analysis is conducted. Figure 10 displays the distribution of the perpendicular magnetic field on the superconducting coils with 3 mm flux divertor. The magnetic field components perpendicular to the surface are greatly reduced in the RPHTSM with the flux divertor, indicating a significant improvement in magnetic leakage when compared to the non-ring structure. However, the maximum perpendicular magnetic field components still occur at the ends of the superconducting coil, which does not show a significant change compared to the excitation coil winding without the flux divertor. The superconducting coils beneath the flux divertor in structures b and c display considerably lesser perpendicular magnetic field components. Subsequently, a fuzzy image preprocessing method based on fuzzy similarities computations will be used to construct classes of images from which to extract a representative fuzzy image, as described in [18], to characterize the improvement of the magnetic field distribution by the flux divertor.

Distribution of perpendicular magnetic field on the superconducting coils with 3 mm flux divertor.
Figure 11 illustrates that the addition of the flux divertor structure leads to a reduced perpendicular magnetic field component in the superconducting strip. For the superconducting excitation coil system equipped with a 3 mm flux divertor structure, the maximum value of the superconducting strip is 0.3955 T. This value is 4.24% lower than the maximum value of the superconducting strip when there is no flux divertor (0.4137 T). Nevertheless, the magnetic induction strength is still in the recommended range of 0.35–0.45 T. However, it should be noted that this flux divertor structure will only provide limited enhancement in machine performance. It is evident that the embedded superconducting coil in structures b and c greatly reduce the maximum perpendicular magnetic field component of the superconducting strip. The perpendicular magnetic field’s value is 0.0692 T and 0.0764 T under structures b and c, respectively, when the flux divertor has a 3 mm thickness. Additionally, the magnetic induction strength is notably reduced in the perpendicular direction.

Perpendicular magnetic field components of superconducting strips with different flux divertors.
Figure 12 illustrates how various excitation system structures affect the critical current of the superconducting strip. Therefore, it is essential to consider alternative excitation system structures to mitigate these drawbacks. Results show that the superconducting strip critical current decays significantly with the use of the superconducting excitation system under the fluxless divertor structure, requiring more turns to ensure the total number of ampere-turns of the flux divertor for machine design is met. This leads to a substantial increase in machine design cost. With the incorporation of the flux divertor structure, the through-current capacity of the superconducting strip has significantly improved. When comparing the three different structure designs, it was observed that structure b and structure c significantly enhanced the critical current capacity of the superconducting strips. It was also observed that the critical current of the superconducting strips increases along with the thickness of the flux divertor. At 4 mm, the superconducting coil excitation system exhibits the strongest current-carrying capability. The critical current of the flux divertor in structure b, is 78. 36 amperes per superconducting strip, which is 44% greater than the critical current of the superconducting strips of the fluxless divertor structure (34.57 A), which is itself 44.12% greater than the critical current of the superconducting strip without the flux divertor structure.

Critical current diagram of superconducting strip with different flux divertors.
In RPHTSM design, the quantity of superconducting strip material used has a direct influence on the manufacturing cost of the machine. Objective assessments state the adoption of a flux divertor may enhance the current-carrying capacity of the superconducting strip and therefore diminish the amount of strip needed. Figure 13 depicts the comparative reductions in strip material consumption due to different flux divertor configurations. The reduction in usage of superconducting strip material is less than 15% with a flux divertor design under structure a, whereas the usage decreased significantly under structures b and c. The material savings were 59.17% and 58.90% for structures b and c correspondingly. Thus, the simulation results demonstrate a decrease in the amount of superconducting strip material utilized compared to the superconducting excitation coil system prior to optimization, regardless of the flux divertor’s structure employed.

Superconducting strip dosage for different flux divertor structures.
The RPHTSM design achieves stationary operation for the superconducting excitation coil by separating the rotor from the inner stator. This innovation diminishes reliance on brushes and rotating dewar in superconducting machines, ultimately elevating machine reliability while streamlining manufacturing processes. Moreover, a detailed magnetic field analysis reveals the intricate and three-dimensional nature of the leaked magnetic field distribution within the machine. To delve deeper into these effects, a comprehensive three-dimensional magnetic field analysis model specific to RPHTSM was established in this study. The simulation findings underscore a significant 67.73% reduction in the critical current of the superconducting coil owing to perpendicular magnetic field effects encountered in the machine. Consequently, this decline necessitates increased usage of strip material.
Our study endeavors to optimize the background magnetic field distribution in the superconducting coil, aiming to enhance its current-carrying capacity. We explore the potential of introducing an innovative flux divertor structure into the RPHTSM superconducting excitation coil system. Our findings reveal that integrating the flux divertor modifies the path of magnetic lines, significantly reducing the perpendicular magnetic field at the extremities of the superconducting coil. As a result, this modification effectively confines the overall perpendicular magnetic field of the entire superconducting strip within a defined range.
Moreover, we investigated the impact of three distinct flux divertor configurations on a range of parameters. Specifically, we focused on the perpendicular magnetic field within the superconducting strip, its critical current decay, and the necessary amount of high-temperature superconducting material. This thorough analysis involved a comprehensive three-dimensional static magnetic field simulation, complemented by experimental measurements of the critical current characteristic curve. Our experimental findings unequivocally illustrate that strategic installation and structural optimization of the flux divertor can significantly enhance the critical current of the superconducting strip by an impressive 44.12%.
Footnotes
Acknowledgements
We express our sincere gratitude to all individuals and institutions who have generously supported and assisted in this study. We are particularly grateful to the National Nature Science Foundation of China (grant no. 52007086), the State Key Laboratory of Reliability and Intelligence of Electrical Equipment (grant no. EERI_KF2021014), Hebei University of Technology and Natural Science Foundation of Nanjing University of Posts and Telecommunications (Grant No. NY223094), for their invaluable support.
