Active shield superconducting magnet optimization method and system

Abstract

The application discloses an active shielding superconducting magnet optimization method and system, the method obtains magnet design requirements and engineering constraints, constructs a coil configuration search space and implements pre-geometry screening; multiple evaluation points are laid out, the magnetic field response of each coil unit current is calculated; the coil working current is solved according to the power supply mode and constraint checking is completed; the DSV magnetic field uniformity, multi-dimensional external leakage magnetic field, coil current-carrying ratio and quench hot spot temperature are calculated synchronously; the manufacturing deviation influence coefficient is introduced to carry out robustness evaluation, and the leakage magnetic grid of the optimized configuration is encrypted for re-evaluation; according to the comprehensive score formula, the configuration is sorted and the design scheme is output. The application realizes the integrated and automatic optimization of the magnet, establishes a multi-dimensional quantitative evaluation system, considers the experimental accuracy and the superconducting operation safety, can effectively reduce the performance attenuation caused by the processing tolerance, the scheme screening is objective and reproducible, and is suitable for the design of active shielding superconducting magnets of a synchrotron radiation light source and an X-ray absorption spectrum device.

Description

Technical Field

[0001] This invention belongs to the field of superconducting magnet design technology, specifically relating to an optimization method and system for actively shielded superconducting magnets. Background Technology

[0002] Superconducting magnets generate high-intensity, stable magnetic fields by carrying large currents through superconducting coils at low temperatures. They are widely used in various beamline experimental devices, including fundamental physics experiments, magnetic characterization of materials, particle beamline manipulation, synchrotron radiation sources, and X-ray absorption spectroscopy. Compared to conventional water-cooled resistance magnets, superconducting magnets have lower operating losses and can operate continuously for extended periods, making them particularly suitable for space-constrained, heat-sensitive in-situ sample characterization systems.

[0003] In experimental stations using synchrotron radiation and X-ray absorption spectroscopy, the core function of a superconducting magnet is to construct a uniform magnetic field of a predetermined intensity in the sample region. This allows the sample to be irradiated by the light beam under strong magnetic field conditions and output quantitatively detectable spectral and diffraction signals. Such scenarios impose multiple rigid constraints on magnet design: First, the effective sample region at the center of the magnet must reach the target magnetic field strength, and the magnetic field uniformity within the sample's bearing area must meet the requirements for quantitative experimental analysis. Second, stray leakage magnetic fields diffused outward from the magnet must not interfere with the light source beamline, detector, sample stage displacement mechanism, cryogenic vacuum components, and surrounding precision electronic instruments. Third, the overall structure of the magnet must provide space for the incident light path, the exit light path, the sample support stage, the cryogenic cooling screen, the refrigerator, and the vacuum shell for assembly and passage; the coil structure must not spatially interfere with the optical path or cryogenic structure.

[0004] To suppress stray fields outside the magnet, existing superconducting magnet shielding solutions are divided into passive shielding and active shielding. Passive shielding involves adding high-permeability ferromagnetic materials around the magnet, relying on the material's magnetic flux absorption and redirection to attenuate leakage magnetic field. While structurally simple, this significantly increases the overall weight and radial volume of the magnet system. In situations where the light source has limited installation space and the optical path requires a through-path, the placement and thickness of the shielding material are easily constrained by space, and the ferromagnetic material is prone to magnetic saturation under high fields, resulting in a significant decrease in shielding effectiveness. Active shielding, on the other hand, adds an independent energized shielding coil outside the main coil. This coil generates a reverse compensating magnetic field to counteract the leakage magnetic field diffused outward from the main coil, resulting in a more compact overall structure, perfectly suited to the layout of compact experimental stations. However, active shielding systems involve multiple windings, including the main coil, axial shielding coil, and outer shielding coil. Strong electromagnetic coupling exists between geometric and current parameters, significantly increasing the number of design variables and making optimization design far more difficult than with passively shielded magnets.

[0005] The complete configuration of an actively shielded superconducting magnet consists of a main coil and at least one type of shielding coil: the main coil is responsible for generating the target magnetic field at the center of the magnet; the axial shielding coil is coaxially arranged at both ends of the main coil, mainly weakening the axial far-field stray magnetic field; the outer shielding coil radially wraps around the outside of the main coil, mainly suppressing leakage magnetic field at the radial equipment boundary; in engineering, axial shielding can be used alone, outer shielding can be used alone, or the two types of shielding coils can be used in combination. Different topologies correspond to completely independent geometric variables, current constraints, and safety verification conditions.

[0006] Currently, the mainstream design method for actively shielded superconducting magnets in the industry still relies on iterative simulations based on engineers' experience: designers pre-determine geometric parameters such as the radius, height, number of layers, and axial spacing of the main coil and shielding coil based on experience, and use finite element magnetic field simulation software to calculate the central magnetic field and external leakage magnetic field; if the central magnetic field does not meet the specifications or the external leakage magnetic field exceeds the limit threshold, the coil size, position, and number of turns are manually adjusted and the simulation calculation is repeated. This manual trial-and-error mode is only suitable for simple magnets with a small number of coils and low variable dimensionality; for composite actively shielded magnets with both axial shielding and outer shielding, the number of design variables increases exponentially, the coupling relationships between variables become complex, and the design efficiency of repeatedly adjusting parameters manually is extremely low, making it difficult to quickly obtain a feasible solution that simultaneously meets all engineering constraints.

[0007] While existing automated design schemes based on intelligent optimization algorithms incorporate heuristic search tools such as genetic algorithms, particle swarm optimization, and Bayesian optimization, using coil geometric parameters as optimization variables and constructing objective functions based on central magnetic field error, external leakage flux, and wire material consumption, six insurmountable engineering shortcomings still exist: (1) Only constrain the magnetic field at the center point, and ignore the uniformity of the magnetic field over the entire sample. Existing optimization schemes only use the magnetic field strength at the geometric center of the magnet as a constraint indicator, without considering that the sample has actual three-dimensional spatial dimensions. In the light source experiment, the sample is not a mathematical single point, but a DSV (Diameter of Spherical Volume) region with the magnetic center as the sphere center. Only ensuring that the magnetic field at the center point meets the standard can easily lead to problems such as excessive peak-to-valley differences in the magnetic field within the spherical sample region and deterioration of magnetic field uniformity, which directly affects the repeatability of experimental data and the accuracy of quantitative analysis. Existing schemes do not incorporate the uniformity of the magnetic field across the entire DSV region into the simultaneous optimization and evaluation process.

[0008] (2) Postponing the superconducting safety verification process and repeatedly reworking the optimized process. The existing design process treats superconducting safety indicators such as quench hotspot temperature, coil critical current margin, and temperature operating margin as post-optimization verification steps: first, multiple candidate coil configurations are screened based on electromagnetic indicators, and then thermal stability and quench protection simulation analyses are conducted separately. If the post-verification finds that the coil hotspot temperature exceeds the standard, the operating current is close to the critical threshold, or the temperature margin is insufficient, all electromagnetic optimization work is invalidated, and it is necessary to return to the front end to readjust the coil geometry and current parameters. This results in a long design iteration cycle and serious waste of resources. Moreover, many intermediate high-quality electromagnetic configurations generated during the optimization process cannot predict superconducting operation risks in advance.

[0009] (3) Manufacturing and assembly deviations were not considered, resulting in poor stability of the project implementation. Active shielding magnets rely on the mutual cancellation of the magnetic field between the main coil and the shielding coil to obtain residual leakage magnetic flux, belonging to a weak margin shielding system of large subtraction. In actual winding, component processing, and assembly, slight manufacturing tolerances will occur in coil height, radial radius, winding layer spacing, and axial installation spacing. Existing optimization simulations are all based on theoretical standard dimensions and do not simulate the changes in leakage magnetic flux after dimensional disturbances. Some configurations with excellent leakage magnetic performance under theoretical dimensions will have their magnetic field cancellation balance disrupted when manufacturing deviations occur, resulting in external leakage magnetic flux exceeding the standard and failing to meet equipment usage requirements.

[0010] (4) There is an inherent contradiction between accuracy and computational efficiency in magnetic flux leakage assessment. External magnetic flux leakage evaluation relies on spatially discrete sampling points: if a high-density three-dimensional mesh is used to arrange the evaluation points, each candidate configuration requires massive magnetic field calculations, resulting in an exponential increase in overall optimization time; if a sparse coarse mesh is used, local magnetic flux leakage peaks within the mesh gaps are easily missed, leading to simulation misjudgments and misclassifying configurations with actual excessive magnetic flux leakage as acceptable. For stringent magnetic flux leakage control standards such as the 5-Gauss line in light source devices, this evaluation flaw directly misleads the design of equipment safety distances and cryogenic structural layouts. Existing technologies cannot simultaneously achieve both high accuracy in magnetic flux leakage evaluation and high overall computational efficiency.

[0011] (5) The space constraints specific to the light source were not included in the optimization screening process in advance. The magnet assembly space of synchrotron radiation and X-ray source experimental stations is subject to multiple exclusive boundary constraints: the incident and exit light paths must have through channels, and the sample stage, cold shield, vacuum shell, and refrigerator occupy fixed installation space. Existing optimization algorithms only set upper and lower limits for the coil's own dimensions, failing to verify during the configuration generation stage whether the coil's outer envelope intrudes into the light path, exceeds the cold shield / vacuum boundary, or exceeds the axial total length limit. Many candidate configurations that meet electromagnetic performance standards are found to have spatial interference problems in the later structural design stage, forcing them to be abandoned entirely and significantly extending the magnet development cycle.

[0012] (6) Multiple evaluation indicators lack standardized comprehensive ranking logic The optimization of active shielding magnets involves more than ten evaluation indicators, such as maximum leakage magnetic field, average leakage magnetic field, root mean square leakage magnetic field, magnetic field uniformity, superconducting safety margin, wire consumption, and magnet volume. Existing technologies only rank the leakage magnetic field values ​​individually and have not established a multi-indicator weighted comprehensive evaluation system. When the leakage magnetic field indicators of multiple candidate configurations are close, it is impossible to automatically prioritize the engineering preferred scheme with low manufacturing sensitivity, sufficient safety margin, and compact structural size. The scheme selection relies on the subjective judgment of engineers, and the reproducibility of the design results is poor.

[0013] Patent publication number CN113889313A discloses a high-field whole-body magnetic resonance imaging active shielding superconducting magnet and its design method. It uses a series-connected integrated structure of a main coil, adjustment coil, and outer shielding coil, and relies on a target field inverse solution algorithm to complete the coil size and current matching design. This can simultaneously constrain the magnetic field uniformity of the central DSV spherical region and the external 5 Gaussian stray field boundary. However, in actual magnet design and application, it is only suitable for large-aperture MRI equipment for human imaging. It lacks a dedicated incident and outgoing optical path avoidance constraint mechanism for synchrotron radiation sources, and is based solely on theoretical standard size simulation calculations, failing to consider the leakage magnetic drift problem caused by coil processing and assembly size deviations. Relying solely on a single magnetic field constraint screening scheme makes it difficult to simultaneously meet the multiple engineering requirements of light source magnet optical path constraint, manufacturing robustness, and superconducting operational safety, thus failing to solve the core challenge of integrated automated optimization of a compact active shielding magnet for synchrotron radiation.

[0014] In summary, existing active shielding superconducting magnet design methods cannot achieve integrated collaborative optimization under multiple constraints, and are particularly unsuitable for the development of compact active shielding superconducting magnets with multiple rigid constraints, such as synchrotron radiation sources and X-ray absorption spectroscopy experimental stations. Therefore, there is an urgent need to provide a novel parameterized collaborative optimization method and supporting system for active shielding superconducting magnets. Summary of the Invention

[0015] The present invention aims to solve the technical problem that existing active shielding superconducting magnet design methods cannot achieve integrated and coordinated optimization under multiple constraints.

[0016] The present invention solves the above-mentioned technical problems through the following technical means: An optimization method for actively shielded superconducting magnets includes the following steps: First, obtain the design requirements of the magnet and the engineering constraints of the optical path and low temperature structure. Then, parameterize the main coil, the optional axial shielding coil, and the optional outer shielding coil into a one-dimensional vector and build a configuration search space. Eliminate invalid configurations that interfere with the coil or intrude into the optical path through geometric pre-screening. Secondly, distinguish the magnetic center point, DSV spherical sampling point, external leakage magnetic field evaluation point, and coil safety evaluation point, and calculate the magnetic field response of each coil per unit current. Secondly, the working current of each coil that satisfies the target center magnetic field is solved by relying on the linear superposition relationship of the magnetic field, and the current and spatial boundary constraints are checked in advance, and the substandard configurations are directly eliminated. Then, for qualified configurations, the magnetic field uniformity of the DSV region, multi-dimensional external leakage flux, coil peak field, current carrying ratio, and quench hot spot temperature are calculated simultaneously. Subsequently, tolerance perturbation was applied to the key coil dimensions to obtain the manufacturing deviation influence coefficient, and the leakage flux local mesh refinement review was carried out on the top-ranked configurations to update the leakage flux index. Finally, a comprehensive score is obtained by multiplying the maximum leakage flux, the manufacturing deviation influence coefficient, and the quench risk penalty coefficient. The magnet geometry, current, and complete set of performance parameters are then output in ascending order of the scores.

[0017] This invention provides a complete solution covering the entire process of magnet design, from configuration construction and magnetic field calculation to multi-dimensional performance assessment, manufacturing robustness verification, and comprehensive optimization. It integrates five major engineering challenges—optical path spatial constraints, sample area uniformity, superconducting safety, leakage magnetic field accuracy, and processing tolerance—into an automated optimization chain. It abandons the fragmented approach of traditional manual step-by-step calculations and item-by-item verification, significantly improving the design efficiency and reliability of active shielded superconducting magnets. This is the core and complete technical solution of this invention.

[0018] Preferably, the one-dimensional vector of the configuration search space is set as Its expression is: ;in Main coil height, Number of main coil layers The inner radius of the main coil, The outer radius of the main coil, The axial distance between the main coil and the axial shielding coil. The height of the axial shielding coil. This refers to the number of axially shielded coil layers. The inner radius of the outer shielding coil. The height of the outer shielding coil. This refers to the number of outer shielding coil layers. This represents the outer radius of the outer shielding coil; based on the user-selected shielding topology, variables corresponding to unused shielding coils are disabled from participating in the search.

[0019] It limits the integrated parameter vector of multiple coils, standardizes all geometric variables of the main coil, axial shield, and outer shield, supports free switching of three types of shield topologies, realizes the standardization and computer recognition of multi-coupled coil structure parameters, avoids logical conflicts of coil size, spacing, inner and outer diameter in advance, reduces invalid search samples, and reduces optimization calculation overhead.

[0020] Preferably, the coil operating current calculation includes two modes: In the series power supply mode, the unified formula for solving the single-channel operating current is as follows: In independent power supply mode, with the central point magnetic field meeting the standard as the equation constraint and the minimum external leakage flux as the optimization objective, the calculation formula is as follows: , In the formula, The total axial magnetic field at the center point under unit current; The magnetic field at the center of the magnet target; Indicates the operating current under series power supply; This represents the row vector consisting of the axial magnetic field response of each coil at the center point under a unit current. Main coil current, For the axially shielded coil current, This refers to the current in the outer shielding coil. This is the current vector in independent power supply mode.

[0021] Simultaneously, the mathematical model for solving the magnetic field is limited to both series and independent power supply modes. By decoupling the coil geometry and current variables through the linear superposition of the magnetic field of unit current, it can not only adapt to low-cost series power supply schemes, but also accurately cancel external stray fields through the independent current ratio of multiple coils, taking into account both equipment cost and leakage magnetic field suppression effect, and adapting to the power supply hardware conditions of different light source devices.

[0022] Preferably, the formula for calculating the uniformity of the DSV magnetic field is: In the formula These represent the maximum and minimum axial magnetic field values ​​within the spherical region of the DSV, respectively. The target is the central magnetic field.

[0023] By introducing a quantification formula for uniformity in the spherical region of DSV, instead of relying solely on the single magnetic field value evaluation scheme at the magnetic center point, the magnetic field consistency of the entire experimental space of the sample is accurately characterized. This avoids the defect of "the central magnetic field meets the standard, but the magnetic field fluctuation in the sample area is too large" from the source, ensuring the repeatability of the light source experimental data and the accuracy of quantitative analysis.

[0024] The preferred formula for calculating multi-dimensional external magnetic flux leakage is as follows: The maximum leakage flux is obtained by traversing all external leakage flux evaluation points. Average leakage flux Root mean square leakage magnetic flux In the formula, This represents the total number of external magnetic flux leakage evaluation points. Indicates the first The leakage flux amplitude at each external leakage flux evaluation point Represents the total axial magnetic field component. Represents the radial magnetic field component. Indicates the maximum leakage flux value. Indicates the average leakage flux value. This represents the root mean square leakage magnetic flux.

[0025] It fully defines three types of leakage magnetic flux quantification calculation methods: single-point leakage magnetic flux, global maximum, average, and root mean square, which multidimensionally characterize the overall distribution of stray field in the magnet. It not only identifies the most dangerous local peak leakage magnetic flux but also evaluates the overall average level of leakage magnetic flux, preventing one-sided screening based on a single indicator and accurately matching the stringent stray field control requirements of the light source beamline and detector.

[0026] Preferably, the formula for calculating the coil current carrying ratio is as follows: In the formula, This is the actual operating current of the coil. This represents the superconducting critical current corresponding to the current peak magnetic field of the coil. The closer it is to 1, the lower the safety margin of the coil operation.

[0027] A quantitative evaluation standard for coil current carrying ratio is added, and the operating margin of the superconducting coil is predicted simultaneously during the optimization and iteration stage. Configurations with high quench risk are screened in advance. This is different from the post-processing mode of existing technologies that only verify safety after optimization. It reduces repeated rework of the scheme and ensures the safety of long-term low-temperature continuous operation of the magnet.

[0028] Preferably, the method for calculating the manufacturing deviation influence coefficient is as follows: ;in The maximum leakage flux under standard theoretical dimensions, The maximum leakage flux is the worst-case operating condition after dimensional perturbation simulation.

[0029] By quantifying the deterioration of leakage magnetic flux caused by manufacturing deviation influence coefficient, the performance fluctuation caused by coil size disturbance is incorporated into the evaluation system. Stable configurations that are not sensitive to processing errors are selected first, thus solving the inherent engineering problem of active shielding magnetic field "large number subtraction and tolerance easily destroying the shielding effect".

[0030] The preferred formula for calculating the comprehensive score is: In the formula, Maximum leakage flux, To determine the influence coefficient of manufacturing deviation, The penalty coefficient for quench risk is used; for candidate configurations with the same score, the auxiliary ranking indicators are leakage flux near the optical path, root mean square value of leakage flux, current carrying ratio, bus material usage, and total axial length of the magnet.

[0031] We construct a weighted comprehensive scoring formula that integrates leakage magnetic field, manufacturing tolerance, and quench risk, and provide supporting multi-dimensional auxiliary ranking rules. This breaks through the limitation of existing technologies that rely solely on leakage magnetic field as a single indicator for selection. It can flexibly balance shielding performance, processing cost, and operational safety according to the actual needs of the light source device. The solution selection logic is standardized and reproducible.

[0032] Preferably, the active shielding superconducting magnet is adapted to a synchrotron radiation source and an X-ray absorption spectroscopy experimental device; external magnetic flux leakage evaluation points are arranged in the outer regions of the incident light path, the outgoing light path, the detector, and the cold screen; the adaptive magnetic flux leakage encryption re-evaluation is only performed on candidate configurations with high comprehensive scores, and a local encryption grid is generated with the global maximum magnetic flux leakage coordinates as the center to update the magnetic flux leakage values.

[0033] By defining specific application scenarios for the light source and rules for arranging magnetic flux leakage evaluation points, and simultaneously clarifying an adaptive magnetic flux leakage encryption mechanism, we can avoid interference problems in the optical path and low-temperature structure in advance. On the other hand, we can eliminate the defect of missing local magnetic flux leakage peaks in the coarse mesh without significantly increasing the global computational load, thus balancing simulation accuracy and optimization speed.

[0034] This invention also discloses a system using the above-described active shielding superconducting magnet optimization method, comprising: The parameter acquisition and candidate configuration generation module is used to read all input constraints, including the target center magnetic field, DSV diameter, optical path position, cold screen boundary, coil size range, wire safety parameters, and manufacturing tolerance range, and to construct parameter vectors, perform pre-geometric screening, and call optimization algorithms to generate legal candidate configurations in batches. The magnetic field response calculation module is used to generate four types of evaluation points and solve the radial and axial magnetic field components per unit current for each coil at all evaluation points. The current calculation and constraint verification module is used to solve the operating current of each coil based on the central magnetic field constraint, and to perform pre-verification of current, geometry, optical path, and low temperature boundary and eliminate unqualified configurations. The performance evaluation module is used to uniformly calculate the DSV magnetic field uniformity, multi-dimensional external leakage flux, coil peak magnetic field, current carrying ratio, temperature margin, and quench hot spot prediction temperature. The Deviation and Encryption Review Module is used to complete the simulation calculation of manufacturing deviation influence coefficient for dimensional disturbance, and to perform local magnetic flux leakage mesh encryption review and update magnetic flux leakage data for high-ranking configurations. The solution output module is used to sort multiple indicators according to the comprehensive score and output recommended configurations, alternative configurations and a complete set of geometric, current and performance parameter reports.

[0035] The advantages of this invention are: (1) Integrated whole process optimization: This invention puts the optical path and cold screen space constraints in advance, and integrates the magnetic field, uniformity, leakage magnetic field and superconducting safety into the same iterative process, and eliminates invalid solutions such as coil interference and current over-limit in advance, saving a lot of repetitive simulations and greatly shortening the development cycle of the light source magnet; (2) Multi-dimensional quantitative evaluation, taking into account both experimental accuracy and equipment protection: abandoning the single evaluation method that relies solely on the magnetic field at the center point and the single-point leakage magnetic field: adopting formulas to quantify the overall magnetic field uniformity of the DSV sample area to ensure the stability of experimental data; at the same time, calculating the maximum, average and root mean square leakage magnetic fields to accurately control the stray fields around the optical path and detector; simultaneously calculating the coil current carrying ratio and quench temperature, eliminating high-risk configurations in advance during the optimization stage, and simultaneously meeting the three requirements of experimental, equipment and coil safety; (3) Balancing computational efficiency and engineering stability: A dedicated mechanism is set up to address the two major pain points of active shielding magnets being sensitive to tolerance and leakage magnetic meshes being prone to missing peak values: the manufacturing deviation coefficient is calculated by dimensional perturbation, and the schemes whose performance is not easily deteriorated after processing are given priority; the leakage magnetic mesh is only locally densified for the top-ranked configurations, without using full-sample high-density simulation, which avoids missing local high stray fields while controlling computational overhead and improving the actual performance of the finished product. (4) Multi-topology standardized design, objective and reproducible scheme selection: The parameter vector is compatible with three coil structures: axial shielding, outer shielding and composite shielding, and can be adapted to different synchrotron radiation and X-ray experimental stations; a weighted comprehensive scoring rule for leakage magnetic field, manufacturing deviation and quench risk is established, and auxiliary indicators such as volume and consumables are matched with the same score, so that it no longer depends on manual subjective selection, and outputs complete coil geometry, current and safety parameters, which is convenient for subsequent structural and cryogenic engineering design. Attached Figure Description

[0036] Figure 1 This is a flowchart of an active shielding superconducting magnet optimization method according to Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the active shielding superconducting magnet structure according to an active shielding superconducting magnet optimization method of Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the candidate configuration parameterization of an active shielding superconducting magnet optimization method according to Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the evaluation point arrangement for an active shielding superconducting magnet optimization method according to Embodiment 1 of the present invention; Figure 5 This is a schematic diagram of the comprehensive evaluation and ranking data flow of an active shielding superconducting magnet optimization method according to Embodiment 1 of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] Example 1: This embodiment provides an optimization method for actively shielded superconducting magnets. To better illustrate the method of this application, the specific structure is first described. (See reference...) Figure 2 The active shielding superconducting magnet includes a main coil for generating a target magnetic field in the sample region. And a shielding coil for reducing external magnetic leakage of the magnet. The shielding coil includes axial shielding coils located on both sides of the main coil. or including an outer shielding coil located radially outside the main coil. or simultaneously include axially shielded coils and outer shielding coil Main coil Arranged around the magnetic center O, these coils provide the primary magnetic field required for the sample region. Axial shielding coils. Arranged in the main coil On both sides of the axis, used to adjust the axial far-field leakage magnetic field of the magnet. Outer shielding coil Arranged radially outside the main coil, it is used to adjust the leakage flux near the radially outside of the magnet and the equipment boundary. As a light source, For reflecting mirrors, These represent incident light and reflected light, respectively. Indicates the sample area. The vacuum enclosure is represented by the magnetic center O, which is located near the sample area. The incident and outgoing light paths pass through the sample area or its vicinity. The space reserved by the cold shield, vacuum enclosure, and cryostat forms the engineering boundary for the arrangement of the magnet. That is, the main coil and shielding coil cannot be arbitrarily arranged solely according to their electromagnetic properties; they also need to avoid the optical path and meet the assembly boundaries of the low-temperature structure. Therefore, the optimization object of this application is not an isolated coil, but rather the overall configuration of the actively shielded superconducting magnet arranged within the experimental space of the light source and the boundary of the low-temperature structure (configuration refers to a complete physical configuration scheme of the actively shielded superconducting magnet).

[0039] See Figure 1 The method includes the following steps: S1, obtaining design requirements and establishing a candidate configuration search space; S2, generating evaluation points and calculating the magnetic field response per unit current; S3, solving for the operating current and performing basic constraint checks; S4, calculating external leakage flux, DSV uniformity, and safety indicators; S5, performing manufacturing deviation stability evaluation and adaptive leakage flux densification review; S6, comprehensively ranking and outputting candidate schemes. For details, please refer to... Figure 1 This embodiment presents an optimization method for actively shielded superconducting magnets, the specific implementation process of which is as follows: S1. Obtain design requirements and establish a candidate configuration search space: Input all constraints such as target magnetic field, sample spherical region, optical path, cold shield boundary, coil size, wire parameters, and manufacturing tolerance; unify the parameterization of main coil, axial shielding coil, and outer shielding coil, supporting three shielding topologies; pre-screen invalid configurations that overlap coils, encroach on the optical path, or exceed the low temperature boundary, and use optimization algorithms to generate candidate schemes in batches.

[0040] The specific process of step S1 includes: First, all engineering input parameters are entered. Based on the actual operating conditions of the synchrotron radiation source or X-ray absorption spectroscopy experimental station, operators batch-enter three types of data: global design parameters, engineering boundary constraint parameters, and optimization algorithm control parameters. The global design parameters include the axial magnetic field of the magnetic center target. The parameters include: the diameter of the DSV spherical region in the sample area, the coordinates of the magnet center, and the external leakage magnetic field control threshold; engineering boundary constraint parameters include the spatial coordinates of the incident and outgoing light paths, the maximum allowable outer dimensions of the cold shield in the radial and axial directions, the vacuum shell boundary, the geometric value ranges of various coils (ranges of inner and outer radii, height, and number of layers of the main coil, the ranges of spacing, height, and number of layers of the axial shielding coil, and the ranges of inner and outer radii, height, and number of layers of the outer shielding coil), the critical performance curve of the superconducting wire, the maximum allowable operating current of the coil, and the preset manufacturing and assembly tolerance range; algorithm control parameters include the total number of optimization iterations, the number of candidate configurations generated in a single iteration, the global sampling ratio, the local neighborhood search shrinkage range, and the convergence judgment threshold.

[0041] Then, a standardized parameter vector is defined to distinguish between the three types of shielded topologies. (See also...) Figure 3 To construct a unified one-dimensional parameter vector: ,in Main coil height, Number of main coil layers The inner radius of the main coil, The outer radius of the main coil, The axial distance between the main coil and the axial shielding coil. The height of the axial shielding coil. This refers to the number of axially shielded coil layers. The inner radius of the outer shielding coil. The height of the outer shielding coil. This refers to the number of outer shielding coil layers. This refers to the outer radius of the outer shielding coil. The system provides three switchable shielding configuration modes: axial shielding coil only, outer shielding coil only, and a combined axial and outer shielding coil. When a single shielding mode is selected, the variables corresponding to the unshielded coils are fixed as invalid values ​​and no longer participate in parameter searching, reducing the dimensionality of the algorithm solution.

[0042] Next, pre-selective geometric screening rules are loaded to establish legal configuration judgment criteria. The system incorporates multi-layer physical constraint screening logic. Any generated parameter vector undergoes rapid geometric verification first. Configurations with the following conditions are directly judged as invalid and removed from the search sample pool: ① The main coil, axial shielding coil, and outer shielding coil overlap spatially, with no insulation assembly gap between windings; ② The outer envelope coordinates of the coil intrude into the incident light path, the outgoing light path, or the sample stage mounting area; ③ The radial dimension of the coil exceeds the radial boundary defined by the cold shield and vacuum shell, and the total axial length exceeds the installation limit of the cryogenic cavity; ④ The inner radius of the outer shielding coil is less than or equal to the outer radius of the main coil, with no reserved space for cooling or support; ⑤ The distance between the axial shielding coil and the main coil is too small, making it impossible to arrange the insulation frame and cooling channel. This screening only performs basic numerical comparisons and does not require magnetic field simulation, resulting in extremely low computational overhead.

[0043] Finally, the optimization algorithm is invoked to generate initial candidate configurations in batches and complete the initial screening. A Bayesian optimization algorithm is used to perform global sampling within the preset value range of each variable, generating multiple sets of initial parameter vectors. Each set of parameter vectors automatically performs the aforementioned preliminary geometric screening. Only valid configurations that fully satisfy all spatial and geometric constraints will proceed to the next step, S2, for magnetic field response calculation. Invalid configurations that are screened out are directly discarded, thus not occupying subsequent simulation calculation resources and reducing redundant calculations without engineering value from the source.

[0044] S2. Generate evaluation points and calculate the magnetic field response per unit current: Divide the points into four categories: magnetic center point, DSV sampling point, external leakage magnetic point, and coil safety point; calculate the magnetic field of all points when each coil carries a unit current, and the total magnetic field of any point is obtained by linearly superimposing the magnetic fields of each coil according to the current.

[0045] The specific process of step S2 includes: First, the evaluation point set includes the center point, DSV sampling points, external leakage magnetic field evaluation points, and safety evaluation points. (See [link / reference]). Figure 4 The magnetic center O is located at the center of the sample area or the field center required by the experiment; a DSV spherical evaluation area is set up with the magnetic center O as the center, and DSV sampling points are arranged in this area; external leakage magnetic evaluation points are arranged outside the magnet, near the incident light path, near the outgoing light path, near the detector, outside the cold screen, or near the equipment boundary; safety evaluation points are arranged near conductors such as the end of the coil, inside the coil, and outside the coil.

[0046] It should be noted that, Figure 4 The DSV sampling points, external leakage flux evaluation points, and safety evaluation points are not the same type of point. DSV sampling points are used to evaluate the magnetic field uniformity within the sample area or effective experimental area; external leakage flux evaluation points are used to evaluate the residual magnetic field in a specified space outside the magnet; and safety evaluation points are used to estimate the peak magnetic field, current carrying capacity, and temperature margin near the coil conductor. Figure 4This application distinguishes between different evaluation points, enabling separate evaluations of the central magnetic field, sample area uniformity, external magnetic leakage, and conductor operational safety, avoiding the conflation of evaluation indicators with different physical purposes into one category. The coordinates of the four evaluation points are calculated analytically based on the input DSV diameter, optical path coordinates, and cold screen boundary parameters. Four sets of independent evaluation points are generated within the axisymmetric coordinate system, and all calculations are based on the axisymmetric cylindrical coordinate system: Magnetic center point: unique reference sampling point with fixed coordinates Specifically used for verifying the central magnetic field of a magnet target. .

[0047] DSV spherical sampling point: defines the radius of the spherical region. , The diameter of the pre-defined DSV spherical region is set; in this embodiment, several sampling points (or other forms of sampling points) are generated on the spherical region centered on the magnetic center. For the axisymmetric model, in According to angle in the cross section The formula for generating sampling points and the coordinates of a single sampling point is: The system calculates and stores the coordinates of all points inside and on the sphere in batches for subsequent calculation of the magnetic field uniformity in the DSV region. and They represent the first Radial and axial coordinates of each DSV sampling point Indicates the first Each sampling angle (to avoid evaluating only the sphere and ignoring variations within the sphere, sampling points can also be generated at several intermediate radii during implementation, for example, using...) (or other proportional radius to generate internal sampling points).

[0048] External magnetic leakage evaluation points: External magnetic leakage evaluation points are used to evaluate the residual magnetic field outside the magnet. These points are generated from points, lines, or regions: when a coordinate point is specified, that point is used as the evaluation point; when a line segment is specified, multiple evaluation points are discretized along the segment; when a region is specified, evaluation points are generated according to radial and axial grids. For light source experimental setups, external magnetic leakage evaluation points are preferentially placed near the incident light path, the exit light path, the detector, the outside of the cold screen, and near the equipment boundary to reflect the actual impact of magnetic leakage on the experimental setup.

[0049] Coil safety evaluation points: These are used to estimate the peak magnetic field near the coil. Safety evaluation points are set at locations such as the inner side of the main coil, the outer side of the main coil, the end of the main coil, the end of the axial shielding coil, and the inner side of the outer shielding coil. The reason for setting safety evaluation points is that the local peak magnetic field experienced by the conductor can affect the critical current and temperature margin of the superconducting wire. If only the central magnetic field and external leakage flux are evaluated, without evaluating the peak field near the conductor, an option with insufficient operating margin may be selected.

[0050] Then, the magnetic field response of each coil under unit current is calculated independently. The magnetic field of the main coil, axial shielding coil and outer shielding coil under the current configuration is calculated separately in sequence. The calculation condition is that a single coil is supplied with a unit current and the current of the other coils is set to zero.

[0051] Definition: The first The coil, the first The unit current magnetic field response vector at each evaluation point is: ;in For the first The coil in the first The axial magnetic field components generated at each evaluation point This represents the radial magnetic field component. If the candidate configuration includes a main coil, an axially shielded coil, and an outer shielded coil, the unit current response of each type of coil is calculated separately. If a certain type of shielded coil is not used, the corresponding current and magnetic field response terms are set to zero, eliminating the need for repeated calculations and saving computational overhead. This step only completes the numerical acquisition of the unit magnetic field of a single coil; the magnetic field value is directly output by the standard electromagnetic field solver module. result.

[0052] Subsequently, a linear superposition relationship of the magnetic fields of multiple coils was established. According to the rules of magnetic field superposition, under any operating current, the first... The total magnetic field at each evaluation point is obtained by linearly superimposing the contributions from each coil:

[0053] For the specific coil in this application, it can be written as: ; Split axial and radial components: ; in Represents the main coil current, Represents the axial shielding coil current, This represents the current of the outer shielding coil. Indicates the first main coil The unit current magnetic field response vector at each evaluation point Indicates the axial shielding coil number The unit current magnetic field response vector at each evaluation point Indicates the outer shielding coil number The unit current magnetic field response vector at each evaluation point. This superposition relationship decouples the coil geometry from the operating current. Step S2 only needs to store the unit magnetic field response based on the fixed geometry, and the subsequent step S3 only needs to adjust the current parameters to quickly update the entire magnetic field, without having to repeatedly recalculate the magnetic field corresponding to the coil geometry, thus greatly reducing the iterative calculation overhead.

[0054] Finally, the categorized cached response data is passed to the downstream S3 step. The magnetic center point is extracted separately. The total axial magnetic field at the center point under unit current is obtained by summing the unit axial magnetic field components of the three types of coils. ; This will be directly used in the current calculation formula for the S3 series power supply mode. The evaluation points are grouped into four categories, and the complete set of axial and radial magnetic field data per unit current corresponding to the center point, DSV sampling point, external leakage flux evaluation point, and coil safety evaluation point is stored respectively. After all the unit magnetic field data of the current complete coil configuration is cached, it automatically switches to S3 to perform the working current solution; it then reads the next combination configuration output by S1 in a loop, completely repeating all point layout, unit magnetic field acquisition, and data caching operations in this step.

[0055] S3. Solve for the working current and perform basic constraint verification: Solve for the coil working current in two power supply modes: series and independent; uniformly verify the upper limit of current, coil geometry, optical path and low temperature space boundary. Configurations that do not meet the constraints are directly eliminated and will not be calculated further.

[0056] The specific process of step S3 includes: First, the coil operating current system is solved under two power supply modes. The corresponding calculation model is selected based on the preset power supply scheme, and the current is solved based on the unit axial magnetic field at the magnetic center point pre-extracted by S2. These modes include series power supply mode and independent power supply mode. (1) Series power supply mode: The main coil, axial shielding coil, and outer shielding coil are connected in series to form a single current loop. All coils share the same operating current, using the formula Solve for the uniform operating current; where, To preset the magnetic field at the center of the magnet target, The total axial magnetic field formed by the superposition of all coils at the magnetic center point under unit current; (2) Independent power supply mode: The main coil, axial shielding coil and outer shielding coil are each configured with independent power supply circuits. The current vector is solved by simultaneously solving the equations with the magnetic field at the magnetic center point meeting the standard as the constraint and the minimum external leakage flux in the whole domain as the optimization objective. The magnetic field constraint condition at the center point is: ,in This represents the row vector consisting of the axial magnetic field response of each coil at the center point under a unit current. This represents the current vector. Meanwhile, the responses at the external leakage flux evaluation points can form a matrix. The goal of current calculation is to satisfy... At the same time, make The overall amplitude is relatively small (in actual implementation, the current solution can be obtained by using constrained quadratic form solution, KKT equation solution, or traversing candidate current combinations).

[0057] Then, the pre-constraint verification is performed in batches, simultaneously verifying four hard constraints for the coil geometry parameters of the current configuration and the obtained operating current. The verification process only involves numerical comparison and does not require repeated magnetic field simulation calculations. ① Current limit constraint: The operating current of each coil must not exceed the preset maximum allowable operating current; if a certain current exceeds the limit, although the configuration may mathematically satisfy the central magnetic field, the actual power supply or wires cannot withstand it and should not be considered as a valid candidate scheme.

[0058] ② Geometric Constraints: Geometric constraints are used to determine whether the radius, height, spacing, and number of layers of each coil are within the allowable range. For outer shielded coils, it is also necessary to determine whether their inner radius is greater than the outer radius of the main coil and to maintain the necessary gap. For axially shielded coils, it is necessary to determine whether they overlap with the main coil and whether they exceed the axial installation boundary; ③ Cold screen and vacuum housing constraints: used to determine whether the outer envelope of the coil exceeds the radial boundary, axial boundary of the cold screen or the space of the vacuum housing; ④ Light source channel constraint: used to determine whether the coil or its outer envelope intrudes into the incident light path, the outgoing light path, or the sample operating space; even if a certain configuration has good leakage magnetic control, it cannot be considered a feasible solution if it occupies the light path.

[0059] ⑤ Low-temperature boundary constraints: Verify the overall radial and axial outer dimensions of the coil to ensure that the overall structure does not exceed the installation boundaries defined by the cold shield and vacuum shell.

[0060] Finally, the qualified configurations are filtered and passed downstream to check the above four constraints one by one. If any one of the checks fails, the configuration is directly determined to be invalid, and all subsequent performance calculations for the configuration are terminated, and it will not enter the S4 step. If all constraint checks are met, the coil geometric parameters, the operating current of each coil, and the unit magnetic field response dataset stored in S2 are packaged together and automatically transferred to the S4 step for simultaneous calculation of DSV magnetic field uniformity, multi-dimensional external leakage flux, and coil superconducting safety related indicators.

[0061] S4. Calculate external leakage flux, DSV uniformity and safety indicators: Statistically measure the magnetic field difference within the effective spherical region of the DSV to characterize uniformity; calculate the external single-point, average and overall leakage flux levels; calculate the coil operating load, temperature margin and quench prediction temperature, and reduce the priority of the scheme if the indicators exceed the standards.

[0062] The specific process of step S4 includes: First, the multi-dimensional external magnetic flux leakage index is calculated by traversing all external magnetic flux leakage evaluation points. The total axial magnetic field at each point is obtained by superimposing the unit magnetic field response and the operating current. Total radial magnetic field Calculate the single-point leakage magnetic amplitude. Based on the leakage magnetic flux amplitude at all locations, three quantitative indicators of leakage magnetic flux were statistically analyzed: Maximum leakage flux ; Average leakage flux ; Root mean square leakage flux .in, This represents the total number of external magnetic flux leakage evaluation points. The maximum magnetic flux leakage value reflects the most unfavorable point, the average magnetic flux leakage value reflects the overall magnetic flux leakage level, and the root mean square magnetic flux leakage (RMS magnetic flux leakage) value reflects the overall intensity of the magnetic flux leakage distribution. For light source experimental setups, evaluation points near the optical path can be selected as key evaluation locations and listed separately in the output results.

[0063] Secondly, the magnetic field uniformity of the DSV spherical region is calculated. Unit current magnetic field response data for all DSV spherical sampling points are retrieved. Combined with the operating current of each coil obtained from S3, the actual axial magnetic field at each sampling point is obtained using the linear superposition formula for the magnetic field. Let the maximum axial magnetic field at the DSV sampling points be... The minimum axial magnetic field is Using formula Calculate the magnetic field inhomogeneity; where, To preset the target center magnetic field, The unit is ppm, and the obtained value characterizes the overall magnetic field fluctuation level of the sample placement area. If the user has set a maximum permissible non-uniformity... Then when If a candidate configuration fails to meet the sample region uniformity requirements, a penalty value is returned or the ranking priority is lowered. It should be noted that DSV uniformity and external magnetic leakage are not the same metric. DSV uniformity reflects the variation of the magnetic field within the sample region, while external magnetic leakage reflects the residual magnetic field outside the magnet. A candidate configuration may have low external magnetic leakage but insufficient sample region uniformity, or it may have good sample region uniformity but high external magnetic leakage. Therefore, this application calculates both separately and considers them simultaneously in the overall ranking.

[0064] Then, based on the magnetic field response at the safety evaluation point, the peak magnetic field near each type of coil is obtained. Combined with the coil operating current Using a conductor critical current model, estimate the critical current. Carrier ratio and temperature margin: (1) Coil current carrying ratio: using the formula Solve; in the formula, This is the actual operating current of the coil. This represents the superconducting critical current corresponding to the current peak magnetic field of the coil. The closer the value is to 1, the closer the operating point is to the conductor's critical capability. Temperature margin reflects the distance between the current operating temperature and the over-critical temperature. If the candidate configuration results in an excessively high peak magnetic field, it may reduce the conductor's critical current, thereby decreasing the temperature margin.

[0065] (2) Simultaneously estimate the quench hotspot temperature of the coil to predict the quench risk level under long-term energized conditions. This estimation is based on the coil inductance. Operating current conductor cross-sectional area Parameters such as detection time and discharge resistance are used to conduct preliminary screening of quench risk during the optimization phase. This hotspot temperature estimation does not replace the final thermo-electric coupling quench analysis, but it can prevent the selection of configurations that are clearly unfavorable to quench protection during the optimization phase.

[0066] Finally, each valid candidate configuration is given a set of evaluation indicators, including maximum leakage magnetic field, average leakage magnetic field, root mean square leakage magnetic field, DSV uniformity, peak magnetic field, current ratio, temperature margin, and hot spot temperature estimate. All indicators calculated in this step are compared with the corresponding thresholds one by one. If any indicator exceeds the preset threshold, the overall ranking priority of the candidate configuration is reduced. After all indicators are calculated, the configuration geometric parameters, current parameters, uniformity, leakage magnetic field, and superconducting safety data are packaged and transferred to step S5.

[0067] S5. Conduct manufacturing deviation stability evaluation and adaptive leakage flux encryption review: Apply tolerance perturbation to the key dimensions of the coil, and compare the simulation results to obtain the influence coefficient of tolerance on leakage flux; only for the top-ranked schemes, encrypt the sampling points in the leakage flux peak area and update the maximum leakage flux value.

[0068] The specific process of step S5 includes: First, in the actual magnet winding and assembly process, the actual positions corresponding to coil height, radius, axial spacing, and number of layers may deviate. If a certain configuration is very sensitive to small dimensional deviations, although its simulation results may be good, increased leakage flux or decreased uniformity may still occur after manufacturing. This invention, based on the design parameters of candidate configurations, applies perturbations to some geometric parameters to form multiple perturbation samples. The simulation of perturbations affecting the critical dimensional tolerances of the coil selects core geometric parameters that affect the shielding effect, including the main coil height. Axial shielding coil height axial spacing Height of outer shielding coil Inner radius of outer shielding coil The above parameters are subjected to random perturbations within a preset machining and assembly tolerance range to generate multiple sets of derived coil geometric models with dimensional deviations. For each perturbed geometric model, the unit current magnetic field response solution (S2), current solution (S3), and leakage flux calculation (S4) are repeatedly executed to obtain the leakage flux amplitude at all external leakage flux evaluation points under the dimensional deviation condition, and the maximum leakage flux under the worst condition in the perturbation simulation is extracted. .

[0069] Secondly, the formula is used to calculate the influence coefficient of manufacturing deviation. Calculate the manufacturing deviation influence coefficient; where, The maximum leakage flux obtained by solving under the standard theoretical dimensions in S4 is... The maximum leakage flux under the worst-case operating condition obtained from the simulation of dimensional tolerance disturbance; The closer the value is to 1, the less sensitive the candidate configuration is to manufacturing deviations; the larger the value, the more prone the candidate configuration is to magnetic leakage degradation under manufacturing deviations. This coefficient serves as a penalty factor in the overall ranking, rather than replacing the original magnetic leakage index.

[0070] Finally, an adaptive magnetic flux leakage (MF) densification review is performed on the candidate configurations. This adaptive local MF densification review only applies to the candidate configurations with the highest overall ranking at the current stage. The coordinate point corresponding to the current global maximum MF leakage amplitude of this configuration is denoted as […]. Local computational intervals are defined radially and axially. Within these intervals, the mesh step size is reduced, and a high-density, refined sampling mesh is generated. For example, in... and Generate grid points within the range ( (These represent the individual grid lengths for each axis); the leakage magnetic flux amplitude at all points within the refined grid is recalculated, and the global maximum leakage magnetic flux value for this configuration is updated by comparison. This avoids missing local high stray magnetic field peaks in coarse grid sampling, thus improving the accuracy of leakage magnetic flux assessment. If a new maximum leakage magnetic flux value is found to be higher than the original evaluation value after the refined re-evaluation, the leakage magnetic flux index of the candidate configuration is updated using the refined re-evaluation results, and it is re-ranked.

[0071] S6. Comprehensive ranking and output of candidate solutions: The comprehensive score is obtained by multiplying the maximum leakage flux, tolerance influence coefficient, and quench penalty coefficient. The higher the score, the better the solution. In case of a tie, auxiliary indicators such as optical path leakage flux, wire consumption, and magnet volume are compared in turn, and a complete set of parameter reports of coil geometry, current, and performance are output.

[0072] The specific process of step S6 includes: First, refer to Figure 5 This step, the comprehensive sorting stage, establishes a correspondence between input parameters, evaluation indicators, and output schemes. Input parameters include the target central magnetic field, DSV diameter, optical path position, cold shield boundary, coil size range, upper current limit, wire parameters, manufacturing deviation, and leakage flux evaluation area. Evaluation indicators include the central magnetic field, maximum external leakage flux, average external leakage flux, root-mean-square external leakage flux, DSV uniformity, peak field, current carrying capacity, temperature margin, estimated hot spot temperature, and manufacturing deviation influence coefficient. Output schemes include recommended configurations, alternative configurations, coil geometric parameters, current parameters, and an indicator table. (See also...) Figure 5 The final recommended scheme in this application is not determined by a single leakage magnetic field value, but is obtained by screening multiple engineering indicators such as central magnetic field, external leakage magnetic field, sample area uniformity, safety margin, manufacturing deviation and initial risk of quenching.

[0073] The comprehensive ranking is not based solely on a single leakage magnetic field value, but rather incorporates external leakage magnetic field, manufacturing deviation impact, quench risk, safety margin, and auxiliary engineering indicators. In this embodiment, the maximum leakage magnetic field value is used as the basic evaluation metric, combined with the manufacturing deviation impact coefficient and the quench risk penalty coefficient to form a comprehensive score. The comprehensive score for candidate configurations is calculated using the following formula. Calculate the overall score; where, This represents the overall score of the candidate configurations. The maximum magnetic flux leakage after local mesh refinement and re-evaluation update. To determine the influence coefficient of manufacturing deviation, This is the penalty coefficient for exceeding the limit. A lower overall score indicates better overall performance of the candidate configuration. If the estimated hotspot temperature of the candidate configuration does not exceed a set threshold, then... The value can be set to 1; if the estimated hotspot temperature exceeds the set threshold, then... A value greater than 1 can be used, thereby lowering the ranking of the candidate configuration.

[0074] Secondly, when the overall scores of several candidate configurations differ slightly, an auxiliary ranking rule is adopted. This auxiliary ranking index includes root-mean-square leakage flux, average leakage flux, manufacturing deviation influence coefficient, hot spot temperature, current carrying capacity, temperature margin, total conductor usage, axial length, radial outer diameter, and leakage flux near the optical path. This auxiliary ranking rule avoids selecting configurations that, while having a slightly lower maximum leakage flux, have a poor overall leakage flux distribution or a low safety margin.

[0075] Finally, the recommended configuration and several alternative configurations are output. Each output configuration includes at least the shielding mode, main coil geometry, axial shielding coil geometry, outer shielding coil geometry, main coil current, axial shielding coil current, outer shielding coil current, center magnetic field, DSV uniformity, maximum leakage flux, average leakage flux, root mean square leakage flux, peak magnetic field, current carrying capacity, temperature margin, estimated hot spot temperature, manufacturing deviation influence coefficient, and comprehensive score. For light source experimental setups, leakage flux values ​​at the incident light path, exit light path, and evaluation points near the detector can also be output, facilitating designers to determine whether the configuration is suitable for a specific experimental station layout.

[0076] The output parameters can directly support the subsequent cryogenic structural design, strength verification, and fabrication drawing preparation of the superconducting magnet. At this point, the single-round active shielding superconducting magnet optimization process is complete.

[0077] Example 2: This embodiment discloses an active shielding superconducting magnet optimization system for executing the aforementioned active shielding superconducting magnet optimization method. The system includes: a parameter acquisition and candidate configuration generation module, a magnetic field response calculation module, a current solution and constraint verification module, a performance evaluation module, a deviation and fine-grained review module, and a scheme output module. These modules work collaboratively to fully realize automated magnet optimization design. The specific implementation process is as follows: Parameter acquisition and candidate configuration generation module: This module reads all input constraints, including the target center magnetic field, DSV diameter, optical path position, cold shield boundary, coil size range, wire safety parameters, and manufacturing tolerance range; constructs a one-dimensional parameter vector of the coil; locks invalid variables according to the preset shielding topology; calls the optimization algorithm to generate candidate configurations in batches; and performs a pre-geometric screening to eliminate invalid configurations that exhibit coil interference, intrude into the optical path, or exceed the low temperature boundary. The selected legal candidate configurations are then sequentially sent to the magnetic field response calculation module.

[0078] The magnetic field response calculation module receives the geometric parameters of valid candidate configurations and classifies them into four types of points: magnetic center point, DSV spherical sampling point, external leakage magnetic field evaluation point, and coil safety evaluation point. It then calculates the axial and radial magnetic field components corresponding to each evaluation point when a unit current is applied to the main coil, axial shielding coil, and outer shielding coil, obtaining the magnetic field response dataset for each group of coils under unit current. Finally, it establishes a linear superposition relationship for the magnetic field, caches all magnetic field response data, and extracts the total axial magnetic field superimposed on the unit current at the magnetic center point. The data is then transmitted to the current solution and constraint verification module.

[0079] Current calculation and constraint verification module: retrieves the unit current magnetic field response data output by the magnetic field response calculation module, and solves for the coil operating current according to the selected power supply mode; the series power supply mode uses the formula The unified operating current is determined. In the independent power supply mode, the objective is to achieve the required magnetic field at the center point and minimize external leakage flux. The independent operating current of each coil is then solved simultaneously. After the current is solved, pre-constraint checks are performed, sequentially verifying the upper limit of the operating current, the range of geometric parameters, the optical path avoidance conditions, and the low-temperature boundary conditions. If any constraint is not met, the current configuration is discarded. Data for qualified configurations that satisfy all constraints are pushed to the performance evaluation module.

[0080] Performance Evaluation Module: Receives qualified configuration geometric parameters, operating current, and unit magnetic field response data, and simultaneously performs quantitative calculations of multiple indicators: It uses the magnetic field data from DSV sampling points to solve for magnetic field inhomogeneity; it traverses external leakage flux evaluation points to calculate leakage flux amplitude, maximum leakage flux, average leakage flux, and root-mean-square leakage flux; it calculates the coil current-carrying ratio based on the magnetic field data from coil safety evaluation points and predicts the overheating hotspot temperature. The complete performance indicator dataset is then transmitted to the deviation and encryption re-evaluation module.

[0081] Deviation and Encryption Review Module: Selects key geometric dimensions of the coil, applies tolerance perturbations, generates multiple sets of derived configurations and completes simulations, extracts the maximum leakage flux under the worst-case operating condition, and applies the formula... Calculate the manufacturing deviation impact coefficient; screen the candidate configurations with the highest overall ranking, establish a local fine mesh centered on the maximum leakage magnetic field coordinate, recalculate the leakage magnetic field and update the maximum leakage magnetic field index; add the manufacturing deviation impact coefficient and the updated leakage magnetic field index to the configuration parameter package and send it to the scheme output module.

[0082] Solution output module: Receives configuration parameter information and uses formulas Calculate a comprehensive score and initially sort the configurations from lowest to highest. For configurations with the same score, conduct a secondary sort based on leakage flux near the optical path, root mean square value of leakage flux, current carrying capacity, bus material usage, and total axial length of the magnet. Based on the sorting results, select recommended and alternative configurations, and output a complete set of performance parameter reports including coil geometry parameters, operating current, magnetic field uniformity, leakage flux index, superconducting safety index, and manufacturing deviation influence coefficient, for use in subsequent structural and cryogenic engineering designs.

[0083] The present invention has the following advantages: This invention provides an optimization method and system for actively shielded superconducting magnets, which prioritizes optical path and cryogenic space constraints, constructing an integrated automated optimization process to overcome the shortcomings of traditional step-by-step design and repeated rework. This application establishes a multi-level quantitative evaluation system, simultaneously calculating DSV magnetic field uniformity, multi-dimensional leakage flux indicators, coil current carrying ratio, and quench risk, overcoming the limitations of single-indicator evaluation. The solution adds a manufacturing deviation influence coefficient assessment and an adaptive leakage flux mesh refinement mechanism, balancing computational efficiency and engineering robustness, and improving the problems of active shielding magnets being sensitive to processing tolerances and easily overlooking local leakage flux peaks. The standardized parameter vector is compatible with various shielding topologies and adaptable to synchrotron radiation sources and X-ray absorption spectroscopy experimental devices; relying on a comprehensive scoring formula combined with multi-level auxiliary sorting rules, the selection logic is objective and reproducible, capable of outputting a complete set of engineering parameters, effectively shortening the superconducting magnet R&D cycle and improving the actual operational stability after the solution is implemented.

[0084] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Terms such as "upper," "lower," "left," "right," "front," and "rear" used in the invention are merely for clarity of description and are not intended to limit the scope of the invention. Changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.

[0085] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An optimization method for actively shielded superconducting magnets, characterized in that, Includes the following steps: First, obtain the design requirements of the magnet and the engineering constraints of the optical path and low temperature structure. Then, parameterize the main coil, the optional axial shielding coil, and the optional outer shielding coil into a one-dimensional vector and build a configuration search space. Eliminate invalid configurations that interfere with the coil or intrude into the optical path through geometric pre-screening. Secondly, distinguish between the magnetic center point, DSV spherical sampling point, external leakage magnetic field evaluation point, and coil safety evaluation point, and calculate the magnetic field response per unit current of each coil. Secondly, the working current of each coil that satisfies the target center magnetic field is solved by relying on the linear superposition relationship of the magnetic field, and the current and spatial boundary constraints are checked in advance, and the substandard configurations are directly eliminated. Then, for qualified configurations, the magnetic field uniformity of the DSV region, multi-dimensional external leakage flux, coil peak field, current carrying ratio, and quench hot spot temperature are calculated simultaneously. Subsequently, tolerance perturbation was applied to the key coil dimensions to obtain the manufacturing deviation influence coefficient, and the leakage flux local mesh refinement review was carried out on the top-ranked configurations to update the leakage flux index. Finally, a comprehensive score is obtained by multiplying the maximum leakage flux, the manufacturing deviation influence coefficient, and the quench risk penalty coefficient. The magnet geometry, current, and complete set of performance parameters are then output in ascending order of the scores.

2. The method for optimizing an actively shielded superconducting magnet according to claim 1, characterized in that, Let the configuration search space be a one-dimensional vector. , ;in Main coil height, Number of main coil layers The inner radius of the main coil, The outer radius of the main coil, The axial distance between the main coil and the axial shielding coil. The height of the axial shielding coil. This refers to the number of axially shielded coil layers. The inner radius of the outer shielding coil. The height of the outer shielding coil. This refers to the number of outer shielding coil layers. This represents the outer radius of the outer shielding coil; based on the user-selected shielding topology, variables corresponding to unused shielding coils are disabled from participating in the search.

3. The method for optimizing an actively shielded superconducting magnet according to claim 1, characterized in that, The calculation of coil operating current includes two modes: In the series power supply mode, the unified formula for solving the single-channel operating current is as follows: In independent power supply mode, with the central point magnetic field meeting the standard as the equation constraint and the minimum external leakage flux as the optimization objective, the calculation formula is as follows: , In the formula, The total axial magnetic field at the center point under unit current; The magnetic field at the center of the magnet target; Indicates the operating current under series power supply; This represents the row vector consisting of the axial magnetic field response of each coil at the center point under a unit current. Main coil current, For the axially shielded coil current, This refers to the current in the outer shielding coil. This is the current vector in independent power supply mode.

4. The method for optimizing an actively shielded superconducting magnet according to claim 1, characterized in that, The formula for calculating the uniformity of the DSV magnetic field is: In the formula These represent the maximum and minimum axial magnetic field values ​​within the spherical region of the DSV, respectively. The target is the central magnetic field.

5. The method for optimizing an actively shielded superconducting magnet according to claim 1, characterized in that, The multi-dimensional external magnetic flux leakage calculation formula is as follows: The maximum leakage flux is obtained by traversing all external leakage flux evaluation points. Average leakage flux Root mean square leakage magnetic flux In the formula, This represents the total number of external magnetic flux leakage evaluation points. Indicates the first The leakage flux amplitude at each external leakage flux evaluation point Represents the total axial magnetic field component. Represents the radial magnetic field component. Indicates the maximum leakage flux value. Indicates the average leakage flux value. This represents the root mean square leakage magnetic flux.

6. The method for optimizing an actively shielded superconducting magnet according to claim 1, characterized in that, The formula for calculating the current carrying ratio of a coil is: In the formula, This is the actual operating current of the coil. This represents the superconducting critical current corresponding to the current peak magnetic field of the coil. The closer it is to 1, the lower the safety margin of the coil operation.

7. The method for optimizing an actively shielded superconducting magnet according to claim 5, characterized in that, The calculation method for the manufacturing deviation influence coefficient is as follows: ;in The maximum leakage flux under standard theoretical dimensions, The maximum leakage flux is the worst-case operating condition after dimensional perturbation simulation.

8. The method for optimizing an actively shielded superconducting magnet according to claim 7, characterized in that, The formula for calculating the overall score is: In the formula, Maximum leakage flux, To determine the influence coefficient of manufacturing deviation, The penalty coefficient for quench risk is used; for candidate configurations with the same score, the auxiliary ranking indicators are leakage flux near the optical path, root mean square value of leakage flux, current carrying ratio, bus material usage, and total axial length of the magnet.

9. The method for optimizing an actively shielded superconducting magnet according to claim 1, characterized in that, The active shielding superconducting magnet is adapted to synchrotron radiation sources and X-ray absorption spectroscopy experimental devices; external magnetic flux leakage evaluation points are arranged in the outer regions of the incident light path, the outgoing light path, the detector, and the cold screen; the adaptive magnetic flux leakage encryption re-evaluation is only performed on candidate configurations with the highest comprehensive scores, and a local encryption grid is generated with the global maximum magnetic flux leakage coordinates as the center to update the magnetic flux leakage values.

10. An active shielding superconducting magnet optimization system, characterized in that, include: The parameter acquisition and candidate configuration generation module is used to read all input constraints, including the target center magnetic field, DSV diameter, optical path position, cold screen boundary, coil size range, wire safety parameters, and manufacturing tolerance range, and to construct parameter vectors, perform pre-geometric screening, and call optimization algorithms to generate legal candidate configurations in batches. The magnetic field response calculation module is used to generate four types of evaluation points and solve the radial and axial magnetic field components per unit current for each coil at all evaluation points. The current calculation and constraint verification module is used to solve the operating current of each coil based on the central magnetic field constraint, and to perform pre-verification of current, geometry, optical path, and low temperature boundary and eliminate unqualified configurations. The performance evaluation module is used to uniformly calculate the DSV magnetic field uniformity, multi-dimensional external leakage flux, coil peak magnetic field, current carrying ratio, temperature margin, and quench hot spot prediction temperature. The Deviation and Encryption Review Module is used to complete the simulation calculation of manufacturing deviation influence coefficient for dimensional disturbance, and to perform local magnetic flux leakage mesh encryption review and update magnetic flux leakage data for high-ranking configurations. The solution output module is used to sort multiple indicators according to the comprehensive score and output recommended configurations, alternative configurations and a complete set of geometric, current and performance parameter reports.

Images