A method, apparatus, device and medium for determining a radiation protection dose of a cyclotron

Abstract

The present application relates to the technical field of radiation protection, and discloses a cyclotron radiation protection dose determination method, device, equipment and medium, the method comprises the following steps: constructing a full-scale geometric model of the cyclotron with a corresponding relationship between the geometric boundary of the beam dynamics calculation domain and the Monte Carlo radiation transport calculation domain; tracking the particle orbit in the beam dynamics environment, detecting and intercepting the six-dimensional phase space state data of the collision particles colliding with the accelerator wall; in the Monte Carlo radiation transport calculation domain, constructing a self-defined radiation source based on the six-dimensional phase space state data; performing particle transport and nuclear reaction simulation calculation using the self-defined radiation source, and outputting the three-dimensional radiation dose field around the cyclotron. Through coupling the beam dynamics simulation and the Monte Carlo radiation transport calculation, the present application can restore the collision state of the particles on the accelerator wall with high fidelity and serve as a radiation source term, thereby improving the accuracy of the radiation protection dose evaluation around the cyclotron.

Description

Technical Field

[0001] This invention belongs to the field of radiation protection technology, specifically relating to a method, apparatus, equipment, and medium for determining radiation protection dose in a cyclotron. Background Technology

[0002] As a core device in fields such as tumor therapy, boron neutron capture therapy (BNCT), and radioisotope production, proton / heavy ion cyclotrons inevitably experience beam loss during operation. When the charged particle beam interacts with components such as the accelerator vacuum chamber walls, electrostatic deflector diaphragms, stripping membranes, or beam-limiting apertures, it generates a mixed radiation field containing high-energy neutrons, gamma rays, and secondary charged particles through nuclear reactions (such as the (p,n) and (p,γ) reactions of protons or the nuclear fragmentation reaction of carbon ions). To ensure the safety of operators and the surrounding environment, and to meet radiation protection regulations and medical physics quality control requirements, it is essential to perform high-precision prediction and assessment of the aforementioned radiation field during the accelerator design and operation phases. This will guide the optimized design of the labyrinth shielding structure and protective facilities.

[0003] Currently, the industry commonly uses Monte Carlo particle transport programs (such as Geant4, MCNP, and FLUKA) for radiation dosimetry calculations. These programs typically employ black-box, macroscopic empirical assumptions. In the Monte Carlo geometric model, it is artificially assumed that beam loss occurs at certain discrete locations within the accelerator, and the lost particles are simplified as isotropic point sources or line sources. Simultaneously, the energy spectrum of the lost particles is often assumed to be a single energy level or follow a simple Gaussian distribution, and the angular distribution is approximated as unidirectional or simple solid-angle emission. This disconnected algorithm based on macroscopic experience ignores the microscopic dynamics of beam loss, leading to significant discrepancies between the calculated results and actual conditions.

[0004] Therefore, a new method for determining radiation protection dose is urgently needed. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method, apparatus, equipment and medium for determining radiation protection dose of a cyclotron accelerator.

[0006] According to one aspect of this application, a method for determining radiation protection dose for a cyclotron is disclosed, the method comprising: Construct a full-scale geometric model of a cyclotron with the geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain; In a beam dynamics environment containing space charge force and radio frequency electromagnetic field, particle trajectories are tracked to detect and acquire colliding particles that collide with the walls of the full-scale geometry of the accelerator. The six-dimensional phase space state data of the colliding particles were intercepted; In the Monte Carlo radiative transport computation domain, a custom radiation source is constructed based on the aforementioned six-dimensional phase space state data; The custom radiation source is used to perform particle transport and nuclear reaction simulation calculations, and the three-dimensional radiation dose field around the cyclotron is output.

[0007] In some embodiments, constructing a full-scale geometric model of a cyclotron with a geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain includes: The triangular facet geometric mesh data used to characterize the three-dimensional geometry of the accelerator vacuum chamber, extraction mechanism and internal components are imported into the geometric model construction process; Control in the geometric model construction process: Based on the triangular facet geometric mesh data, a beam dynamics computational domain and a Monte Carlo radiation transport computational domain are constructed respectively. The beam dynamics computational domain and the Monte Carlo radiation transport computational domain share the same spatial coordinate origin, coordinate axis direction, and spatial scale to unify their spatial coordinate reference. During the construction process, the geometric boundary in the beam dynamics computational domain used for particle-wall collision is mapped to the starting boundary in the Monte Carlo radiation transport computational domain used for setting radiation source terms, so that the spatial coordinates of the colliding particles intercepted in the beam dynamics computational domain are used as the coordinates of the radiation source terms in the Monte Carlo radiation transport computational domain. Based on the triangular facet geometric mesh data, a spatial layered bounding structure is constructed. The spatial layered bounding structure is used to group the triangular facets according to their spatial positions and wrap them layer by layer into nested bounding boxes to accelerate the detection of collisions between particle trajectories and geometric surfaces during particle trajectory tracking, thereby obtaining the full-scale geometric model of the cyclotron.

[0008] In some embodiments, tracking particle trajectories in a beam dynamics environment containing space charge force and radio frequency electromagnetic field to detect and acquire colliding particles that collide with the walls of the full-scale geometry of the accelerator includes: In the beam dynamics computational domain, physical field data characterizing the accelerator magnetic field distribution, radio frequency electric field distribution, and beam space charge effect are loaded to construct a beam dynamics environment that includes space charge force and radio frequency electromagnetic field. A higher-order symplectic integral algorithm is used to calculate the propulsion of the particle trajectory in the beam dynamics environment, and the trajectory of the particle in the current step is parameterized as a spatial ray in each integration time step. Based on the spatial layered enclosure structure in the full-scale geometric model, the intersection detection between the spatial ray and the triangular facet geometric mesh data in the full-scale geometric model is performed to determine whether the particle collides with the wall of the full-scale geometric model within the integration time step. The particles that collide are identified as colliding particles.

[0009] In some embodiments, intercepting the six-dimensional phase space state data of the colliding particles includes: Freeze the Hamiltonian dynamics evolution of the colliding particles in the beam dynamics environment; The spatial coordinates, momentum direction, and kinetic energy of the colliding particle at the moment of collision are extracted, and the spatial coordinates, momentum direction, and kinetic energy are defined as the six-dimensional phase space state data of the colliding particle.

[0010] In some embodiments, constructing a custom radiation source based on the six-dimensional phase space state data in the Monte Carlo radiation transport computation domain includes: Obtain the macroscopic weighting factors associated with the colliding particles; Read the six-dimensional phase space state data of the captured colliding particles; A discrete probability distribution model is constructed based on the macroscopic weighting factors to characterize the probability that each collision particle is extracted as the initial particle in the Monte Carlo simulation. In response to the initialization request of the Monte Carlo procedure, sampling is performed according to the discrete probability distribution model, and the spatial coordinates, momentum direction and kinetic energy of the sampled colliding particles are mapped to the starting position, emission direction and particle energy of the radiation source term in the Monte Carlo radiation transport calculation domain. The starting position, emission direction, and particle energy of the radiation source term are used as custom radiation sources.

[0011] In some embodiments, the simulation calculation of particle transport and nuclear reaction using the custom radiation source, outputting a three-dimensional radiation dose field around the cyclotron, includes: In the Monte Carlo radiation transport computation domain, a kernel database covering the cyclotron wall material, shielding structure material, and air medium is loaded; The hadron cascade and evaporation physics model is invoked, and based on the macroscopic weighting factor and the particle energy and emission direction carried in the custom radiation source, the transport calculation of particle-matter interaction is performed. The transport calculation includes nuclear cascade reaction, nuclear evaporation reaction and secondary particle transport process. A three-dimensional spatial grid scorer is established within a preset dose assessment area to cumulatively record the energy deposition and particle flux generated during the transport calculation process; Based on the cumulative results of the three-dimensional spatial grid scorer, and combined with the preset dose conversion coefficient, the three-dimensional distribution field of the ambient dose equivalent around the cyclotron is calculated and used as the three-dimensional radiation dose field.

[0012] In some embodiments, the method further includes: Based on the vector relationship between the starting position and the collision point position of the colliding particle within the current integration time step, the momentum direction and kinetic energy of the colliding particle at the collision point are reversed by interpolation or back-pushing correction, so that the corrected momentum direction and kinetic energy approximate the actual physical state of the colliding particle at the wall collision interface. The macroscopic weighting factor corresponding to the colliding particle in the beam dynamics simulation is extracted, and the six-dimensional phase space state data of the colliding particle is associated and stored with the macroscopic weighting factor. The six-dimensional phase space state data includes the corrected spatial coordinates, momentum direction and kinetic energy.

[0013] According to another aspect of this application, a cyclotron radiation protection dose determination device is also disclosed, the device comprising: The cyclotron full-scale geometric model construction module is used to construct a cyclotron full-scale geometric model with the geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain. The collision particle detection module is used to track particle trajectories in a beam dynamics environment containing space charge force and radio frequency electromagnetic field in order to detect and acquire collision particles that collide with the walls of the full-scale geometric model of the accelerator. A six-dimensional phase space state data acquisition module is used to capture the six-dimensional phase space state data of the colliding particles; A custom radiation source construction module is used to construct a custom radiation source based on the six-dimensional phase space state data in the Monte Carlo radiation transport computation domain. The three-dimensional radiation dose field determination module is used to perform particle transport and nuclear reaction simulation calculations using the custom radiation source and output the three-dimensional radiation dose field around the cyclotron.

[0014] According to another aspect of this application, an electronic device is also disclosed, the electronic device including a memory and at least one processor, the memory storing instructions; the at least one processor invokes the instructions in the memory to cause the electronic device to perform the steps of the cyclotron radiation protection dose determination method as described in any of the preceding claims.

[0015] According to another aspect of this application, a computer-readable storage medium is also disclosed, wherein instructions are stored on the computer-readable storage medium, characterized in that, when executed by a processor, the instructions implement the various steps of the cyclotron radiation protection dose determination method as described in any of the preceding claims.

[0016] The present invention includes, but is not limited to, the following beneficial effects: (1) This scheme tracks particle trajectories in a beam dynamics environment containing space charge force and radio frequency electromagnetic field, and intercepts the six-dimensional phase space state data of colliding particles, so that the initial source terms of subsequent radiation transport are no longer artificially assumed ideal distributions, but directly derived from the actual beam dynamics behavior of the accelerator, thereby improving the accuracy of the position, energy and angle correlation of radiation source terms; (2) By constructing a full-scale geometric model with the geometric boundary correspondence between the beam dynamics calculation domain and the Monte Carlo radiation transport calculation domain, and making the two calculation domains share the same spatial coordinate origin, coordinate axis pointing and spatial scale, it avoids In the traditional separate modeling method, common problems such as boundary misalignment, coordinate offset or scale inconsistency can be solved by directly mapping the coordinates of the collision particles intercepted in the beam dynamics to the starting position of the radiation source term in the Monte Carlo computational domain, so that the location of beam loss corresponds spatially to the location from which radiation starts transport; (3) In the full-scale geometric model, a spatial layered enclosing structure can be constructed based on the triangular facet geometric mesh data, which accelerates the detection of trajectory and wall collisions during particle trajectory tracking; at the same time, the Monte Carlo side can rely on this consistent geometric boundary to perform accurate nuclear reactions and secondary particle transport, reducing the computational cost of exhaustive collision detection and making three-dimensional dose assessment engineering feasible. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0018] Figure 1 This is a flowchart of the method for determining radiation protection dose of a cyclotron according to an embodiment of this application; Figure 2 This is a structural block diagram of the cyclotron radiation protection dose determination device according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0019] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] For ease of understanding, the specific process of the embodiments of the present invention will be described below. Figure 1 This is a flowchart of the method for determining the radiation protection dose of a cyclotron according to an embodiment of this application. (See attached document.) Figure 1 It includes the following steps: S100. Construct a full-scale geometric model of the cyclotron with the geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain.

[0021] For example, step S100 can be implemented based on the following steps: The triangular mesh data used to characterize the three-dimensional geometry of the accelerator vacuum chamber, extraction mechanism, and internal components is imported into the geometric model construction process.

[0022] Specifically, firstly, three-dimensional geometric information is acquired to characterize the overall structure of the cyclotron. This three-dimensional geometric information includes at least the accelerator vacuum chamber, extraction mechanism, internal components, and other physical structures that may affect particle orbits or radiation transport paths.

[0023] In practical implementation, the aforementioned three-dimensional geometric information can be provided in the form of triangular facet geometric mesh data. For example, by discretizing the CAD model of a cyclotron, a surface mesh composed of a large number of triangular facets can be obtained. Each triangular facet can be defined by the spatial coordinates of its three vertices, its facet normal vector, and the identification information of the component it belongs to. This triangular facet geometric mesh data can accurately describe the spatial morphology of complex curved surfaces such as the inner wall of the vacuum chamber, the wall of the exhaust pipe, and the outer surface of internal components.

[0024] Subsequently, the aforementioned triangular facet geometric mesh data is imported into the geometric model construction process, serving as a common geometric data source for the subsequent construction of the beam dynamics computational domain and the Monte Carlo radiation transport computational domain.

[0025] Further controls are performed during the geometry model building process: Based on triangular mesh data, a beam dynamics computational domain and a Monte Carlo radiation transport computational domain are constructed respectively. The beam dynamics computational domain and the Monte Carlo radiation transport computational domain share the same spatial coordinate origin, coordinate axis direction, and spatial scale to unify their spatial coordinate reference. In the construction process, the geometric boundary in the beam dynamics computational domain used for particle-wall collision is mapped to the initial boundary in the Monte Carlo radiation transport computational domain used for setting radiation source terms, so that the spatial coordinates of the colliding particles intercepted in the beam dynamics computational domain are used as the coordinates of the radiation source terms in the Monte Carlo radiation transport computational domain. A spatially layered bounding structure is constructed based on the triangular facet geometric mesh data. This structure is used to group the triangular facets according to their spatial location and wrap them layer by layer into nested bounding boxes. This accelerates the detection of collisions between particle trajectories and geometric surfaces during particle trajectory tracking, resulting in a full-scale geometric model of the cyclotron.

[0026] The beam dynamics computational domain is primarily used to track the trajectories of charged particles in an environment containing space charge forces and radio frequency electromagnetic fields, and to detect whether particles collide with the accelerator walls or the surfaces of internal components. Therefore, this beam dynamics computational domain must at least include wall boundary information that can participate in collision detection, such as the inner walls of the vacuum chamber, the walls of the extraction channels, and the surfaces of the beam-limiting elements.

[0027] The Monte Carlo radiation transport computation domain is used to perform particle transport, nuclear reaction, and secondary particle propagation calculations in subsequent steps. Therefore, in addition to the accelerator's internal structure, this Monte Carlo radiation transport computation domain may further include the surrounding shielding structure, building space, air medium region, and dose assessment region.

[0028] It should be noted that although the two computational domains have different uses and may have different mesh or geometric representations, for example, the beam dynamics side can be represented by triangular facets / enclosing structures suitable for fast collision detection, while the Monte Carlo side can be represented by solid construction geometry CSG, voxel mesh, or tetrahedral / hexahedral mesh, both are derived from the same set of triangular facet geometric mesh data, thus maintaining the consistency of key geometric features.

[0029] Furthermore, during the construction process, the beam dynamics computational domain and the Monte Carlo radiation transport computational domain share the same spatial coordinate origin, the same coordinate axis orientation, and the same spatial scale to unify their spatial coordinate references.

[0030] Specifically, the accelerator center, the main magnetic field center, or a mechanical mounting reference point can be used as a common coordinate origin; the X, Y, and Z axes should maintain consistent orientation in both computational domains; and the units of length should uniformly adopt the same physical dimensions such as millimeters, centimeters, or meters. In this way, when the spatial coordinates of a colliding particle are obtained in the beam dynamics computational domain, these coordinates can directly correspond to the spatial position in the Monte Carlo radiation transport computational domain without undergoing complex coordinate rotation, translation, or scaling transformations.

[0031] This approach avoids the coordinate offset, scale error, or boundary misalignment problems caused by separate modeling and calibration of two sets of models in traditional methods.

[0032] Furthermore, during the construction process, the geometric boundaries used in the beam dynamics computational domain to determine particle-wall collisions are mapped to the initial boundaries used in the Monte Carlo radiative transport computational domain to set the radiation source terms. That is, in the beam dynamics simulation, the wall, inner wall panel, or section of the exit pipe wall that a particle is detected colliding with should have the same or essentially identical geometric surface at the corresponding location in the Monte Carlo model. This surface serves as both the boundary where beam loss occurs and the initial emission boundary for subsequent custom radiation sources. Therefore, the spatial coordinates of the colliding particles intercepted in the beam dynamics computational domain can be directly used as the coordinates of the radiation source terms in the Monte Carlo radiative transport computational domain. Similarly, the incident direction, kinetic energy, or momentum direction and kinetic energy after inverse interpolation correction of the colliding particles at this boundary can also be used as the emission direction and particle energy of the custom radiation source. This boundary mapping relationship is applicable to scenarios such as vacuum chamber wall loss, exit mechanism wall loss, and beam-limiting structure edge-rubbing loss. For example, when beam dynamics tracking results show that a particle grazes and collides at a triangular facet of the extraction pipe, the same triangular facet position in the Monte Carlo model can be set as the radiation source starting facet of the lost particle, thus making the generation positions of secondary neutrons, photons, and charged particles closer to the real physical process.

[0033] Furthermore, to improve the efficiency of wall collision detection during particle trajectory tracking, a spatially layered bounding structure can be constructed based on the triangular facet geometric mesh data. Specifically, triangular facets can be grouped according to their spatial location in the triangular facet geometric mesh data, and then wrapped layer by layer into nested bounding boxes according to spatial hierarchy. For example, a hierarchical bounding box tree (BVH), octree, kd-tree, or similar spatial index structure can be constructed. The bottom-level leaf nodes correspond to the actual triangular facets or a small set of facets, while the upper-level nodes correspond to progressively coarser bounding boxes. During particle trajectory tracking, it is first determined whether the particle's trajectory within the current integration step intersects with the higher-level bounding box; if they do not intersect, there is no need to check each fine facet inside the bounding box; if they intersect, the process continues to perform more refined intersection checks on the lower-level subdivision structure until it is determined whether the trajectory collides with a specific triangular facet.

[0034] The spatial layered enclosure structure described above can significantly reduce the computational cost of collision detection between particle trajectories and geometric surfaces. Especially in complex geometric models at all scales, it can effectively avoid exhaustive traversal of all triangular facets, thereby improving beam dynamics tracking efficiency.

[0035] Finally, based on the aforementioned unified coordinate benchmark, shared geometric data source, boundary mapping relationships, and spatially layered enclosing structure, a full-scale geometric model of the cyclotron is obtained. This full-scale geometric model of the cyclotron does not merely refer to a complete geometric replication in a single mesh form, but rather to a comprehensive geometric framework that maintains geometric consistency and boundary correspondence between beam dynamics calculations and Monte Carlo radiation transport calculations. With the help of this full-scale cyclotron model, subsequent steps can complete the entire process of beam loss detection—phase space state interception—custom radiation source construction—radiation dose field output within the same physical space, thereby improving the accuracy and three-dimensional spatial resolution of radiation protection dose assessment around the cyclotron.

[0036] S102. In a beam dynamics environment containing space charge force and radio frequency electromagnetic field, track particle trajectories to detect and acquire colliding particles that collide with the walls of the full-scale geometric model of the accelerator.

[0037] For example, step S102 may include the following steps: First, in the beam dynamics computational domain, physical field data characterizing the accelerator magnetic field distribution, radio frequency electric field distribution, and beam space charge effect are loaded to construct a beam dynamics environment that includes space charge force and radio frequency electromagnetic field.

[0038] The magnetic field distribution can include the main magnetic field, correction magnetic field, extraction magnetic field, or local gradient field, which are used to determine the cyclotron radius, axial focusing / defocusing characteristics, and phase motion behavior of charged particles in the cyclotron. The radio frequency electric field distribution can be used to describe the high-frequency acceleration / deflection field in the cavity or extraction structure, which changes with time or particle phase. The beam's own space charge effect is used to characterize the self-field force generated between macroparticles due to charge repulsion under high current conditions. This effect will change the probability of lateral particle dispersion, longitudinal focusing, and edge particle loss.

[0039] For example, the aforementioned physical fields can be loaded into the computational domain in the form of a grid field diagram, analytical functions, interpolation tables, or Poisson / electromagnetic field solution results. During each integration step or each lap of tracking, the system can read or interpolate the corresponding Lorentz force term and space charge force term based on the particle's current spatial position and phase, thus ensuring that the particle's motion equations are driven not only by the external electromagnetic field but also by the beam's internal self-field.

[0040] Secondly, a higher-order symplectic integral algorithm is used to calculate the propulsion of the particle trajectory in the beam dynamics environment, and the trajectory of the particle in the current step is parameterized as a spatial ray within each integration time step.

[0041] Because particle motion in a cyclotron exhibits periodicity, long-term accumulation, and phase dependence, conventional explicit integration methods can easily introduce artificial energy drift or phase errors. Therefore, symplectic integration algorithms, such as the Sinusoidal method, leapfrog method, higher-order Runge-Kutta symplectic methods, and time-split symplectic integrators, are preferred to better preserve the phase space volume and energy-phase structure of the Hamiltonian system, thereby improving the stability of long-term trajectory tracking. Within each integration time step, the net force or equivalent acceleration acting on the particle can be calculated based on its current state variables, including position, momentum or velocity, energy, and phase, and the particle's state at the next moment can be predicted.

[0042] Furthermore, based on the spatial layered enclosure structure in the full-scale geometric model, intersection detection is performed between the spatial ray and the triangular facet geometric mesh data in the full-scale geometric model to determine whether the particle collides with the wall of the full-scale geometric model within the integration time step.

[0043] Specifically, a coarse screening can be performed first using spatial hierarchical bounding box structures such as the aforementioned hierarchical bounding box tree, octree, and kd-tree. If the spatial ray corresponding to the current integration step does not intersect with a higher-level bounding box, the triangular facets contained within that bounding box are skipped; if the spatial ray enters the range of a bounding box, the detection continues down to the next lower-level node until a leaf node containing the actual triangular facet is reached. For triangular facets that require fine detection, the intersection relationship between the spatial ray and the plane containing the triangular facet, as well as the facet boundary, can be calculated.

[0044] Ultimately, the particles that collided were identified as colliding particles.

[0045] Collisions include not only the initial contact between the particle trajectory and the inner wall of the vacuum chamber, the wall of the extraction pipe, the beam-limiting element, or the surface of internal components, but also beam loss scenarios such as edge rubbing losses, pre-contact scattering near the wall, and boundary intersections before penetrating extremely thin structures. For individual particles identified as colliding particles, their collision-related information can be recorded or temporarily stored, such as the spatial coordinates of the collision, the triangular facet number corresponding to the collision point, the momentum direction before / after the collision, kinetic energy or energy, incident angle, macroscopic weighting factor, current integration step size information, and the macroparticle identifier.

[0046] For example, if a macroparticle moves from its starting position to a collision point where it intersects with a triangular facet of the wall within the current step size, then the macroparticle can be marked as a collision particle.

[0047] S104. Intercept the six-dimensional phase space state data of the colliding particles.

[0048] Specifically, step S104 may include the following steps: First, the Hamiltonian dynamics evolution of frozen colliding particles in a beam dynamics environment.

[0049] Specifically, freezing refers to terminating subsequent orbital integration and electromagnetic field propagation for the colliding particle at the numerical simulation level. When a particle is determined to have collided with the wall of the full-scale geometry model of a cyclotron, it no longer participates in the cyclotron acceleration process as a beam particle confined in the acceleration field, but instead transforms into a potential primary radiation source term. At this point, continuing to track the complex interactions of the particle after penetrating the wall in the beam dynamics computational domain would not only increase unnecessary computational overhead but may also lead to redundant calculations or physical description mismatches due to the different applicability of the two physical models (beam dynamics model and radiation transport model). By freezing the Hamiltonian dynamics evolution of the particle, its state variables at the instant of collision can be fixed, allowing it to be extracted as a complete phase space sample or macroparticle. For Hamiltonian systems described using symplectic integral algorithms, freezing means preserving the particle's set of regular variables in phase space and no longer performing time-dependent evolution propagation calculations on it.

[0050] Furthermore, the spatial coordinates, momentum direction, and kinetic energy of the colliding particles at the moment of collision are extracted, and the spatial coordinates, momentum direction, and kinetic energy are defined as the six-dimensional phase space state data of the colliding particles.

[0051] Here, the spatial coordinates correspond to the collision point positions determined by the intersection detection of spatial rays and triangular facets in the aforementioned steps, or the actual collision interface coordinates of the wall obtained after inverse interpolation / backtracking correction. Under a unified spatial coordinate reference, these coordinates can be directly mapped to the radiation source term positions in the Monte Carlo radiative transport computation domain. Momentum direction refers to the direction of particle motion at the instant of impact with the wall, usually determined by the direction of its momentum vector. This direction determines the initial emission directions of primary and secondary particles in subsequent Monte Carlo simulations. Kinetic energy refers to the kinetic energy possessed by the particle at the moment of collision, which can be determined by its momentum magnitude and rest mass.

[0052] Defining this set of data as six-dimensional phase space state data and capturing it means preserving the crucial phase space correlation information of colliding particles at the moment of beam loss, avoiding the physical information collapse caused by the decoupling approximation of position, energy, and direction in traditional point source models. The captured six-dimensional phase space state data can be directly used for the construction of custom radiation sources in subsequent steps, enabling Monte Carlo radiation transport calculations to simulate particle transport and nuclear reactions based on high-fidelity beam loss phase space distribution, thereby improving the calculation accuracy of the three-dimensional radiation dose field around the cyclotron.

[0053] S106. In the Monte Carlo radiation transport computation domain, a custom radiation source is constructed based on the six-dimensional phase space state data.

[0054] For example, step S106 may include the following steps: First, obtain the macroscopic weighting factors associated with colliding particles.

[0055] In beam dynamics simulations, due to computational resource constraints, macro-particle or super-particle methods are typically used to sample and represent the real beam, meaning one simulated particle can represent several real physical particles. Therefore, each captured collision particle, in addition to its own six-dimensional phase space state, is associated with a corresponding macro-weighting factor. This macro-weighting factor can be used to characterize the number of real particles equivalent to the corresponding macro-particle, its relative contribution to beam intensity, or its proportion of flux in the total beam loss.

[0056] For example, if a macroparticle undergoes space charge field correction, radio frequency phase selection, or particle splitting / merging operations during tracking, its weighting factor can be dynamically adjusted accordingly. Obtaining this factor when constructing the radiation source helps maintain quantitative consistency between the radiation source intensity and the actual beam loss rate in subsequent Monte Carlo simulations, avoiding an overall overestimation or underestimation of the dose rate due to simple equal weighting.

[0057] Secondly, the six-dimensional phase space state data of the captured colliding particles are read.

[0058] Since the beam dynamics computational domain and the Monte Carlo radiative transport computational domain share the same spatial coordinate origin, coordinate axis orientation, and spatial scale, the data read here can be directly applied to the corresponding positions in the Monte Carlo geometric model without complex coordinate transformations. This reading operation preserves the distribution characteristics and correlations of each colliding particle in phase space, such as energy differences at different spatial locations and the forward angle distribution of near-wall rubbing losses, providing a raw data foundation for subsequent high-fidelity source term construction.

[0059] Furthermore, a discrete probability distribution model is constructed based on macroscopic weighting factors to characterize the probability that each colliding particle is extracted as an initial particle in the Monte Carlo simulation. Since beam dynamics simulations may capture a large number of collision particles (or macroparticle samples), and Monte Carlo programs typically perform statistical analysis on a certain number of emitted history particles (Histories) during each initialization, it is necessary to establish a sampling rule from the captured samples to the initial Monte Carlo particles. For example, suppose a total of N collision particle samples are captured, and the macroscopic weighting factor associated with the i-th sample is... Then the following discrete probability distribution model can be constructed: ; In the formula, Let represent the probability that the i-th collision particle sample is selected as the initial particle in the Monte Carlo simulation. The larger the weight of the sample, the greater its contribution to the actual beam loss, and the higher its probability of being selected in the Monte Carlo sampling.

[0060] Furthermore, in response to the initialization request of the Monte Carlo procedure, sampling is performed according to the discrete probability distribution model, and the spatial coordinates, momentum direction and kinetic energy of the sampled colliding particles are mapped to the starting position, emission direction and particle energy of the radiation source term in the Monte Carlo radiation transport computation domain.

[0061] When the Monte Carlo radiation transport procedure starts or enters a new round of historical particle initialization, the program sends an initialization request to the custom source interface. At this time, the system performs random sampling based on the aforementioned discrete probability distribution model, such as using inverse transform sampling, the alias method, or table lookup interpolation, to select a target collision particle sample from the intercepted collision particle samples. For the k-th target collision particle sample, its six-dimensional phase space state data can be mapped as follows: The spatial coordinates of the target collision particle sample are mapped to the starting position of the Monte Carlo radiation source term; The momentum direction of the target collision particle sample is mapped to the emission direction of the initial Monte Carlo particle; The kinetic energy of the target collision particle sample is mapped to the particle energy of the initial Monte Carlo particle.

[0062] Ultimately, the starting position, emission direction, and particle energy of the radiation source term are used as the custom radiation source.

[0063] After the above sampling and mapping operations, the Monte Carlo program no longer relies on traditional isotropic point sources, fixed energy line sources, or empirically assumed sources. Instead, it uses the phase space distribution of colliding particles obtained from beam dynamics tracking as the initial condition. This custom radiation source is physically closer to the actual beam loss process that occurs in a cyclotron, and can reflect the spatial non-uniformity of the loss location, the angular correlation of the incident direction, and the phase space covariance characteristics of the kinetic energy.

[0064] The custom radiation source constructed in this way can be directly used for subsequent particle transport and nuclear reaction simulation calculations, enabling the Monte Carlo program to more accurately simulate the nuclear reactions of primary particles at the wall, the generation and propagation of secondary neutrons and photons, and improve the computational fidelity and spatial resolution of the three-dimensional radiation dose field around the cyclotron.

[0065] S108. Perform particle transport and nuclear reaction simulation calculations using a custom radiation source, and output a three-dimensional radiation dose field around the cyclotron.

[0066] For example, step S108 may include the following steps: First, in the Monte Carlo radiation transport computational domain, a kernel database covering the cyclotron wall material, shielding structure material, and air medium is loaded.

[0067] Specifically, in Monte Carlo particle transport simulations, the nuclear database provides essential data such as cross-sectional data, angular distribution data, energy allocation data, and secondary particle yield information necessary for the interaction of particles with matter at different energies. For cyclotron applications, the nuclear database covers nuclear reaction data corresponding to the accelerator vacuum chamber wall materials (e.g., stainless steel, aluminum, copper, or titanium alloys), extraction mechanism and beam-limiting element materials, external shielding structure materials (e.g., ordinary concrete, heavy concrete, boron-containing polyethylene, lead, or iron), and the air medium. For example, the nuclear database may include Monte Carlo program formats processed based on evaluation nuclear databases such as ENDF, JEFF, JENDL, and CENDL, and may also include extended sets of data suitable for hadron transport in the high-energy region. By loading the nuclear database corresponding to the aforementioned materials, the corresponding microscopic cross-section and reaction channel information can be accurately retrieved during subsequent transport processes, regardless of the geometric region to which primary or secondary particles propagate.

[0068] Secondly, the hadron cascade and evaporation physics model is invoked, and based on the macroscopic weighting factor and the particle energy and emission direction carried in the custom radiation source, the transport calculation of particle-matter interaction is performed. The transport calculation includes nuclear cascade reactions, nuclear evaporation reactions and secondary particle transport processes.

[0069] Because beam-loss particles in cyclotron accelerators can possess high energies, their interaction with wall or shielding materials can lead not only to conventional electromagnetic energy deposition but also to complex nuclear reactions. Therefore, the Hadron Cascade Model and Evaporation Model are preferred for describing high-energy nuclear cascades, pre-equilibrium emission, recombination evaporation, and residual nuclear de-excitation. In transport calculations, each initial Monte Carlo particle in the custom radiation source carries its initial position, emission direction, particle energy, and corresponding macroscopic weighting factors. The simulation program progressively advances the particle's motion within the geometric model based on this initial state. When a particle interacts with the nuclei of the medium, the system uses the nuclear database and the aforementioned physical models to determine whether elastic scattering, inelastic scattering, nuclear fragmentation, capture reactions, or evaporation reactions occur, generating secondary neutrons, photons, protons, alpha particles, or other charged particles. Transport calculations continuously track the free path, collisions, energy loss, and regeneration processes of these secondary particles until their energy falls below the cutoff energy, they leave the region of interest, or they are absorbed.

[0070] Through the above process, the mixed radiation field generated by the beam loss of the cyclotron can be accurately reflected, especially key phenomena such as the forward peak transport of high-energy neutrons, multiple scattering in the shield, and the generation of trapped gamma rays.

[0071] Furthermore, a three-dimensional spatial grid scorer is established within the preset dose assessment area to cumulatively record the energy deposition and particle flux generated during the transport calculation process.

[0072] To obtain a dose distribution with spatial resolution, rather than just a total count at a single point, a three-dimensional spatial grid can be defined within a pre-defined dose assessment area, such as around the accelerator, on the outer surface of the shield, in maze-like passages, in the operating room, treatment room, or adjacent rooms. This grid can be implemented using a Cartesian grid, cylindrical coordinate grid, voxel grid, or a scorer / tally structure based on geometric units. During particle transport, whenever a particle passes through a grid unit, undergoes energy deposition within a grid unit, or experiences a nuclear reaction, the scorer accumulates and records the corresponding unit based on the particle's weighting factor.

[0073] Finally, based on the cumulative results of the three-dimensional spatial grid scorer and combined with the preset dose conversion coefficient, the three-dimensional distribution field of the ambient dose equivalent around the cyclotron is calculated as the three-dimensional radiation dose field.

[0074] The energy deposition or particle flux obtained directly from Monte Carlo transport typically still needs to be converted into dosimetric quantities usable in radiation protection. For example, this can be based on the energy deposition density in the grid cells, combined with the dielectric mass density. Converted to absorbed dose: ; In the formula, The energy deposited by the scorer, where V is the corresponding grid cell volume. The density is the medium.

[0075] For neutron, photon, and mixed radiation fields, preset dose conversion coefficients can be used to convert particle flux / flux into ambient dose equivalent H∗(10), effective dose, or tissue equivalent dose, etc., as radiation protection quantities. For example, coefficients based on ICRP, ICRU recommended data, or built-in conversion tables can be used to weighted superposition of the flux contributions of neutrons or photons of different energies.

[0076] Furthermore, Figure 2 This is a structural block diagram of a cyclotron radiation protection dose determination device according to an embodiment of the application, as shown below. Figure 2 As shown, the device includes: The cyclotron full-scale geometric model construction module is used to construct a cyclotron full-scale geometric model with the geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain. The collision particle detection module is used to track particle trajectories in a beam dynamics environment containing space charge force and radio frequency electromagnetic field in order to detect and acquire collision particles that collide with the walls of the full-scale geometric model of the accelerator. The six-dimensional phase space state data acquisition module is used to capture the six-dimensional phase space state data of colliding particles; A custom radiation source construction module is used to construct custom radiation sources based on six-dimensional phase space state data in the Monte Carlo radiation transport computation domain. The three-dimensional radiation dose field determination module is used to perform particle transport and nuclear reaction simulation calculations using a custom radiation source and output the three-dimensional radiation dose field around the cyclotron.

[0077] The application of the relevant modules of the system in this example can be found in the above introduction to the principles of the method, and will not be repeated here.

[0078] above Figure 2 The present invention provides a detailed description of a cyclotron radiation protection dose determination device from the perspective of modular functional entities. The electronic equipment in the present invention is described in detail below from the perspective of hardware processing.

[0079] Figure 3 This is a schematic diagram of the structure of an electronic device 300 provided in an embodiment of the present invention. The electronic device 300 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 310 (e.g., one or more processors) and a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) for storing application programs 333 or data 332. The memory 320 and storage media 330 can be temporary or persistent storage. The program stored in the storage media 330 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the electronic device 300. Furthermore, the processor 310 may be configured to communicate with the storage media 330 and execute the series of instruction operations in the storage media 330 on the electronic device 300.

[0080] Electronic device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Server, MacOSX, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The illustrated electronic device structure does not constitute a limitation on electronic devices and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0081] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of any of the above-described methods for determining radiation protection dose for a cyclotron.

[0082] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0083] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0084] 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. A method for determining the radiation protection dose of a cyclotron accelerator, characterized in that, The method includes: Construct a full-scale geometric model of a cyclotron with the geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain; In a beam dynamics environment containing space charge force and radio frequency electromagnetic field, particle trajectories are tracked to detect and acquire colliding particles that collide with the walls of the full-scale geometry of the accelerator. The six-dimensional phase space state data of the colliding particles were intercepted; In the Monte Carlo radiative transport computation domain, a custom radiation source is constructed based on the aforementioned six-dimensional phase space state data; The custom radiation source is used to perform particle transport and nuclear reaction simulation calculations, and the three-dimensional radiation dose field around the cyclotron is output.

2. The method for determining radiation protection dose for a cyclotron according to claim 1, characterized in that, The construction of the full-scale geometric model of the cyclotron with the geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain includes: The triangular facet geometric mesh data used to characterize the three-dimensional geometry of the accelerator vacuum chamber, extraction mechanism and internal components are imported into the geometric model construction process; Control in the geometric model construction process: Based on the triangular facet geometric mesh data, a beam dynamics computational domain and a Monte Carlo radiation transport computational domain are constructed respectively. The beam dynamics computational domain and the Monte Carlo radiation transport computational domain share the same spatial coordinate origin, coordinate axis direction, and spatial scale to unify their spatial coordinate reference. During the construction process, the geometric boundary in the beam dynamics computational domain used for particle-wall collision is mapped to the starting boundary in the Monte Carlo radiation transport computational domain used for setting radiation source terms, so that the spatial coordinates of the colliding particles intercepted in the beam dynamics computational domain are used as the coordinates of the radiation source terms in the Monte Carlo radiation transport computational domain. Based on the triangular facet geometric mesh data, a spatial layered bounding structure is constructed. The spatial layered bounding structure is used to group the triangular facets according to their spatial positions and wrap them layer by layer into nested bounding boxes to accelerate the detection of collisions between particle trajectories and geometric surfaces during particle trajectory tracking, thereby obtaining the full-scale geometric model of the cyclotron.

3. The method for determining radiation protection dose for a cyclotron according to claim 1, characterized in that, The process of tracking particle trajectories in a beam dynamics environment containing space charge force and radio frequency electromagnetic field to detect and acquire colliding particles that collide with the walls of the full-scale geometry of the accelerator includes: In the beam dynamics computational domain, physical field data characterizing the accelerator magnetic field distribution, radio frequency electric field distribution, and beam space charge effect are loaded to construct a beam dynamics environment that includes space charge force and radio frequency electromagnetic field. A higher-order symplectic integral algorithm is used to calculate the propulsion of the particle trajectory in the beam dynamics environment, and the trajectory of the particle in the current step is parameterized as a spatial ray in each integration time step. Based on the spatial layered enclosure structure in the full-scale geometric model, the intersection detection between the spatial ray and the triangular facet geometric mesh data in the full-scale geometric model is performed to determine whether the particle collides with the wall of the full-scale geometric model within the integration time step. The particles that collide are identified as colliding particles.

4. The method for determining radiation protection dose for a cyclotron according to claim 1, characterized in that, The intercepted six-dimensional phase space state data of the colliding particles includes: Freeze the Hamiltonian dynamics evolution of the colliding particles in the beam dynamics environment; The spatial coordinates, momentum direction, and kinetic energy of the colliding particle at the moment of collision are extracted, and the spatial coordinates, momentum direction, and kinetic energy are defined as the six-dimensional phase space state data of the colliding particle.

5. The method for determining radiation protection dose for a cyclotron according to claim 4, characterized in that, The construction of a custom radiation source based on the six-dimensional phase space state data in the Monte Carlo radiation transport computation domain includes: Obtain the macroscopic weighting factors associated with the colliding particles; Read the six-dimensional phase space state data of the captured colliding particles; A discrete probability distribution model is constructed based on the macroscopic weighting factors to characterize the probability that each collision particle is extracted as the initial particle in the Monte Carlo simulation. In response to the initialization request of the Monte Carlo procedure, sampling is performed according to the discrete probability distribution model, and the spatial coordinates, momentum direction and kinetic energy of the sampled colliding particles are mapped to the starting position, emission direction and particle energy of the radiation source term in the Monte Carlo radiation transport computation domain. The starting position, emission direction, and particle energy of the radiation source term are used as custom radiation sources.

6. The method for determining radiation protection dose for a cyclotron according to claim 1, characterized in that, The simulation calculations of particle transport and nuclear reaction using the custom radiation source output a three-dimensional radiation dose field around the cyclotron accelerator, including: In the Monte Carlo radiation transport computation domain, a kernel database covering the cyclotron wall material, shielding structure material, and air medium is loaded; The hadron cascade and evaporation physics model is invoked, and based on the macroscopic weighting factor and the particle energy and emission direction carried in the custom radiation source, the transport calculation of particle-matter interaction is performed. The transport calculation includes nuclear cascade reaction, nuclear evaporation reaction and secondary particle transport process. A three-dimensional spatial grid scorer is established within a preset dose assessment area to cumulatively record the energy deposition and particle flux generated during the transport calculation process; Based on the cumulative results of the three-dimensional spatial grid scorer, and combined with the preset dose conversion coefficient, the three-dimensional distribution field of the ambient dose equivalent around the cyclotron is calculated and used as the three-dimensional radiation dose field.

7. The method for determining radiation protection dose for a cyclotron according to claim 4, characterized in that, The method further includes: Based on the vector relationship between the starting position and the collision point position of the colliding particle within the current integration time step, the momentum direction and kinetic energy of the colliding particle at the collision point are reversed by interpolation or back-pushing correction, so that the corrected momentum direction and kinetic energy approximate the actual physical state of the colliding particle at the wall collision interface. The macroscopic weighting factor corresponding to the colliding particle in the beam dynamics simulation is extracted, and the six-dimensional phase space state data of the colliding particle is associated and stored with the macroscopic weighting factor. The six-dimensional phase space state data includes the corrected spatial coordinates, momentum direction and kinetic energy.

8. A device for determining radiation protection dose in a cyclotron, characterized in that, The device includes: The cyclotron full-scale geometric model construction module is used to construct a cyclotron full-scale geometric model with the geometric boundary correspondence between the beam dynamics computational domain and the Monte Carlo radiation transport computational domain. The collision particle detection module is used to track particle trajectories in a beam dynamics environment containing space charge force and radio frequency electromagnetic field in order to detect and acquire collision particles that collide with the walls of the full-scale geometric model of the accelerator. A six-dimensional phase space state data acquisition module is used to capture the six-dimensional phase space state data of the colliding particles; A custom radiation source construction module is used to construct a custom radiation source based on the six-dimensional phase space state data in the Monte Carlo radiation transport computation domain. The three-dimensional radiation dose field determination module is used to perform particle transport and nuclear reaction simulation calculations using the custom radiation source and output the three-dimensional radiation dose field around the cyclotron.

9. An electronic device, characterized in that, The electronic device includes a memory and at least one processor, the memory storing instructions; the at least one processor invokes the instructions in the memory to cause the electronic device to perform the steps of the cyclotron radiation protection dose determination method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the various steps of the method for determining radiation protection dose of a cyclotron as described in any one of claims 1-7.

Images