Method and apparatus for generating ballast model

Abstract

The present application relates to the technical field of digital simulation, and provides a ballast model generation method and device. The ballast model generation method comprises: determining a particle size grading standard of ballast; obtaining a size range of the ballast based on the particle size grading standard; generating a random convex hull in the size range; and performing different scale recess processing on at least two face sheets of the random convex hull respectively to generate a ballast model. The particle size grading standard determines the size distribution characteristics of the ballast particle group, the particle size grading standard is converted into the size range of the ballast, the distribution requirement of the group level can be decomposed into the generation constraint of the individual particle, the random convex hull generated thereby provides a basic grid structure for subsequent multi-scale recess processing, the recess processing with different scales is applied to the random convex hull, the multi-level roughness and irregularity can be introduced, so that the roughness and irregularity of the real ballast particle surface are simulated, and the accuracy of the simulation result based on the ballast model in reflecting the real mechanical performance of the ballast layer is improved.

Description

Technical Field

[0001] This invention relates to the field of digital simulation technology, and provides a method and apparatus for generating ballast models. Background Technology

[0002] Ballast, as a fundamental component of railway track bed, ensures the long-term safe operation of the railway system through multiple functions such as transmitting train loads, absorbing vibrations, and drainage. The shape, size, and distribution of ballast directly affect the load-bearing capacity, stability, and durability of railway tracks. Therefore, accurately reflecting the physical characteristics of ballast has become an important topic in railway engineering simulation research.

[0003] In existing ballast modeling studies, a 3D scan of a small number of typical ballast samples is typically used to obtain the ballast model. However, ballast in actual working conditions is formed from broken rock, and its geometric features are highly irregular and vary from person to person. This modeling method of 3D scanning a small number of typical ballast samples can only obtain a limited number of fixed-shape samples, which cannot fully simulate and reproduce the randomness and diversity of ballast morphology. As a result, the simulation results are difficult to meet the accuracy requirements in reflecting the true mechanical properties of the ballast layer. Summary of the Invention

[0004] This invention provides a method and apparatus for generating ballast models to address the shortcomings of related technologies in fully simulating and reproducing the randomness and diversity of ballast morphology, and to improve the accuracy of simulation results based on ballast models in reflecting the true mechanical properties of the ballast layer.

[0005] This invention provides a method for generating a ballast model, comprising: Determine the particle size distribution standard for ballast; The size range of the ballast is obtained based on the particle size distribution standard; Generate a random convex hull within the specified size range; At least two faces of the random convex hull are concave at different scales to generate a ballast model.

[0006] According to one embodiment of the present invention, obtaining the size range of the ballast based on the particle size distribution standard includes: At least two gradation intervals are obtained based on the particle size distribution standard; The size range of the ballast is obtained by introducing random disturbances into each of the gradation intervals.

[0007] According to an embodiment of the present invention, generating a random convex hull within the stated size range includes: An intermediate polyhedron is generated within the specified size range; If the intermediate polyhedron can pass through the sieve holes, the random convex hull is obtained based on the intermediate polyhedron; wherein, the gradation interval quality is within the set quality range of the gradation interval, and the gradation interval quality is obtained by accumulating the quality of the random convex hull.

[0008] According to one embodiment of the present invention, the intermediate polyhedron is capable of passing through a sieve, comprising: The sieve aperture size range is determined based on the gradation interval corresponding to the size range; If the area of ​​the target facet of the intermediate polyhedron is within the range of the sieve aperture size, then the intermediate polyhedron can pass through the sieve aperture.

[0009] According to one embodiment of the present invention, it further includes: Calculate the target index of the intermediate polyhedron; the target index includes at least one of the needle-like index and the sheet-like index; If the target index is within the set index range, then the intermediate polyhedron is determined to be the target polyhedron; wherein, the ratio of the constraint mass to the gradation interval mass is within the set ratio range, and the constraint mass is obtained by accumulating the mass of the target polyhedron.

[0010] According to one embodiment of the present invention, the size range of the ballast is determined based on its length, width, and height; The step of performing concave processing on at least two faces of the random convex hull at different scales includes: The average size is determined based on the length, the width, and the height; The average size is subjected to a first weighting process to obtain a small indentation parameter, and the average size is subjected to a second weighting process to obtain a large indentation parameter; Based on the small depression parameter and the large depression parameter, at least two facets of the random convex hull are subjected to depression processing of different scales.

[0011] According to one embodiment of the present invention, the small depression parameters include the small depression radius and the small depression depth; The random convex hull surface is recessed according to the small recess parameter, including: The first target patch is randomly determined; Based on the fact that the depth of the small indentation decreases with the square of the distance, the vertices within the radius of the small indentation at the center of the first target facet are moved to perform indentation processing on the facet of the random convex hull.

[0012] According to one embodiment of the present invention, the large depression parameters include the large depression radius and the large depression depth; The random convex hull surface is concave according to the large concavity parameter, including: A second target patch is randomly determined; wherein, the distance between the centers of adjacent second target patches is not less than the interference distance, and the interference distance is obtained by a third weighting process based on the average size; Based on the fact that the depth of the large depression decreases with the cube of the distance, the vertices within the radius of the large depression at the center of the first target facet are moved to perform a depression process on the facet of the random convex hull.

[0013] According to one embodiment of the present invention, after generating the ballast model, the method further includes: At least one of the model file and quality information of the ballast model is stored in a gradation range directory corresponding to the particle size distribution standard, based on the size range of the ballast model.

[0014] This invention provides an apparatus for generating ballast models, comprising: The particle size distribution standard determination module is used to determine the particle size distribution standard of ballast. A size range determination module is used to determine the size range of the ballast based on the particle size distribution standard. A random convex hull generation module is used to generate a random convex hull within the stated size range; The ballast model generation module is used to perform concave processing on at least two facets of the random convex hull at different scales to generate a ballast model.

[0015] This invention provides a method and apparatus for generating a ballast model. By determining the particle size distribution standard of the ballast, the size range of the ballast is obtained based on the particle size distribution standard. Random convex hulls are generated within the size range, and at least two facets of the random convex hulls are subjected to concave processing of different scales to generate a ballast model. This method can simulate the roughness and irregularity of the surface of real ballast particles, thereby improving the accuracy of simulation results based on the ballast model in reflecting the real mechanical properties of the ballast layer. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of the method for generating ballast models provided by the present invention.

[0018] Figure 2 This is a schematic diagram of the gradation curve of the ballast model generation method provided by the present invention.

[0019] Figure 3 It is a schematic diagram of the actual ballast.

[0020] Figure 4 This is a schematic topographical diagram of a ballast model generated by the ballast model generation method provided by the present invention.

[0021] Figure 5 This is a schematic structural diagram of the ballast model generation device provided by the present invention.

[0022] Figure 6 This is a schematic diagram of the physical structure of the electronic device provided by the present invention. Detailed Implementation

[0023] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0024] As mentioned earlier, accurately reflecting the physical characteristics of ballast through ballast models is an important topic in railway engineering simulation research. However, currently, when ballast models are obtained by 3D scanning a small number of typical ballast samples, the resulting ballast models generally require manual adjustment of the ballast mass distribution to meet gradation requirements.

[0025] Furthermore, the difficulty in modeling ballast in actual practice stems from its complex physical properties. Obtaining ballast models through 3D scanning of a small number of typical ballast samples fails to capture the randomness of the ballast's shape and spatial distribution, resulting in models with insufficient representativeness. Factors such as the irregular polyhedral shape, wide grain size distribution, and surface roughness of ballast all affect the mechanical properties of the ballast layer. Currently, ballast models obtained through 3D scanning of a small number of typical samples are insufficient to simulate these complex interactions, leading to poor reliability of simulation results obtained using the large-scale discrete element method.

[0026] Based on this, such as Figures 1 to 6 As shown, the present invention provides a method and apparatus for generating ballast models.

[0027] Figure 1 This is a flowchart illustrating the method for generating ballast models provided by the present invention, as shown below. Figure 1 As shown, the method includes the following steps.

[0028] Step 101: Determine the particle size distribution standard for ballast.

[0029] Ballast, also known as ballast particles or ballast crushed stone, is coarse gravel or crushed stone used to pave the subgrade of highways or railways.

[0030] Particle size distribution standards are used to characterize the required distribution ratio of ballast particles within different particle size ranges. For example, a particle size distribution standard may include the aperture size of a square-hole sieve. In some embodiments, the particle size distribution standard may directly include parameters such as the corresponding percentage of mass passing through the sieve.

[0031] It should be noted that there are many ways to determine the particle size distribution standard for ballast, such as through existing national standards, industry standards, or specific engineering projects for railway crushed stone ballast. Specific methods can be found in relevant specifications or directly specified based on engineering requirements. This embodiment does not impose any limitations on this. For example, the particle size distribution standard can be the requirements for extra-grade crushed stone ballast or first-grade crushed stone ballast.

[0032] Understandably, particle size distribution standards can serve as a benchmark for the subsequent generation of ballast models, determining the size distribution characteristics of ballast particle groups and providing constraints for the modeling process of ballast particles.

[0033] Step 102: Obtain the size range of the ballast based on the particle size distribution standard.

[0034] The ballast size range refers to the limits of the geometric dimensions of ballast particles in three-dimensional space. For example, the ballast size range may include the minimum and maximum allowable values ​​for the length, width, and height of each particle, to define the geometric constraint range of a single ballast in the length, width, and height directions.

[0035] It should be noted that the size range obtained based on the particle size distribution standard can be mapped to the corresponding three-dimensional particle size range by mapping the sieve aperture size specified in the particle size distribution standard. Based on this, there are many ways to determine the size range of ballast according to the three-dimensional particle size range, such as using the interval median, interval endpoints, or introducing appropriate perturbation offsets, etc. This embodiment does not limit the specific methods used.

[0036] It is understandable that by converting the gradation standard into the size range of ballast, the distribution requirements at the group level can be decomposed into the generation constraints of individual particles, providing a specific parameter space for the subsequent generation of random convex hulls.

[0037] Step 103: Generate a random convex hull within the specified size range.

[0038] The random convex hull is a geometric model of a convex polyhedron used to represent the initial basic shape of the ballast. For example, the random convex hull can be composed of a randomly generated set of vertices, all of which lie outside the surface of the random convex hull.

[0039] It should be noted that there are many ways to generate this vertex set. For example, when the size range constrains the coordinate range of the vertices, the vertex set can be generated by uniform distribution random or Gaussian perturbation random. In this embodiment, there is no limitation on this, as long as the generated random convex hull falls within the size range.

[0040] Understandably, the random convex hull provides the initial basic shape of the ballast and can provide the basic mesh structure for subsequent multi-scale concave processing.

[0041] Step 104: Perform concave processing on at least two faces of the random convex hull at different scales to generate a ballast model.

[0042] Among them, the facets of the random convex hull can be triangular facets, polygonal facets, etc.

[0043] Depression processing refers to applying an inward vertex displacement operation to a facet of a random convex hull, causing the facet to naturally stretch and reconstruct, forming a continuous concave geometry on the surface of the random convex hull. Different scales refer to variations in parameters such as the radius and depth of the depression operation.

[0044] It should be noted that applying different scales of concavity treatment to different surfaces can include setting different concavity radii, depths, and shape parameters. For example, some surfaces can be concave in a small area to simulate surface roughness, while other surfaces can be concave in a large area to simulate fractured surfaces or cracks. The specific concavity parameters, the selection of the affected surfaces, and the type of concavity function can be set according to the ballast characteristics to be simulated; this embodiment does not impose any limitations on these settings.

[0045] Understandably, by applying concave treatments of different scales to different facets, multi-layered roughness and irregularity can be introduced, significantly enhancing the surface irregularity and geometric realism of the ballast model, making the final generated random ballast model closer to the shape of real crushed stone.

[0046] The ballast model generation method provided in this embodiment of the invention determines the particle size distribution standard of the ballast, obtains the size range of the ballast based on the particle size distribution standard, generates a random convex hull within the size range, and performs concave processing on at least two facets of the random convex hull at different scales to generate a ballast model. This method can simulate the roughness and irregularity of the surface of real ballast particles, thereby improving the accuracy of simulation results based on the ballast model in reflecting the real mechanical properties of the ballast layer.

[0047] Furthermore, by performing concave processing on at least two facets of the random convex hull at different scales, it is possible to superimpose pit structures of varying sizes on the surface of the ballast particle model, effectively enhancing the roughness and geometric irregularity of the ballast model, thereby significantly improving its realistic reflection of complex mechanical behaviors such as contact, friction, and rolling in discrete element simulation.

[0048] Furthermore, the generated ballast model meets the standard gradation requirements, which reduces the difficulty of screening and eliminating unqualified ballast models in the later stage and improves the generation efficiency of ballast particle groups.

[0049] Based on the above embodiments, obtaining the size range of the ballast based on the particle size distribution standard includes: At least two gradation intervals are obtained based on the particle size distribution standard; The size range of the ballast is obtained by introducing random disturbances into each of the gradation intervals.

[0050] The gradation interval is a particle size range obtained by dividing the particle size gradation standard according to the sieve aperture size. In some embodiments, each gradation interval corresponds to a set of upper and lower sieve aperture values ​​and a sieve passing mass percentage requirement.

[0051] Random perturbation is a random offset applied to the ballast dimensions.

[0052] It should be noted that introducing random perturbation can be done by adding a random offset to the upper and lower limits of each gradation interval, thereby reducing the regularity of the final generated ballast model.

[0053] The amplitude of the disturbance can be adjusted by a preset coefficient, such as a small Gaussian disturbance, but this embodiment does not limit this. The preset coefficient can be set based on the width ratio of the gradation interval or engineering experience, but this embodiment does not limit this.

[0054] It is understandable that by introducing random perturbations on the basis of the gradation interval, the size of the generated random convex hull can be continuously distributed, ensuring that the final generated ballast model meets the constraints of the gradation specification while more closely resembling the natural fluctuation characteristics of the actual ballast particle size.

[0055] Based on any of the above embodiments, generating a random convex hull within the size range includes: An intermediate polyhedron is generated within the specified size range; If the intermediate polyhedron can pass through the sieve holes, the random convex hull is obtained based on the intermediate polyhedron; wherein, the gradation interval quality is within the set quality range of the gradation interval, and the gradation interval quality is obtained by accumulating the quality of the random convex hull.

[0056] Among them, intermediate polyhedra refer to candidate particles randomly generated within a size range.

[0057] Understandably, if the intermediate polyhedron can pass through the sieve, a random convex hull can be obtained based on the candidate particle; if the intermediate polyhedron cannot pass through the sieve, the candidate particle is discarded directly.

[0058] The gradation interval mass, also known as the gradation interval weight, refers to the cumulative mass of all particles that have passed the sieve inspection within the gradation interval corresponding to the size range of the intermediate polyhedron. The gradation interval set mass range refers to the upper and lower limits of the cumulative mass obtained based on the percentage of sieve mass corresponding to the particle size gradation standard.

[0059] It is understandable that if the intermediate polyhedron can pass through the sieve, then the mass of the gradation interval is the cumulative mass value after accumulating the mass of the candidate particle.

[0060] For example, for each gradation interval, after generating the intermediate polyhedron, the sieve aperture passability standard can be determined first based on the sieve aperture size range of the gradation interval. Based on the sieve aperture passability standard, the sieve aperture passability is judged. If the intermediate polyhedron can pass through the sieve aperture, the mass of the intermediate polyhedron is calculated and added to the current total mass of the gradation interval to obtain the gradation interval mass. If the gradation interval mass does not exceed the set mass of the gradation interval, the intermediate polyhedron is determined to be a random convex hull, and the above generation and judgment process is repeated until the gradation interval mass falls within the set mass range of the gradation interval.

[0061] Each gradation interval has a set threshold for its set quality, which can also be called the target quality of the gradation interval. It can be understood that if the total quality of the gradation interval, after being accumulated from the mass of the intermediate polyhedrons, exceeds the set quality threshold, then the intermediate polyhedron is discarded.

[0062] The target quality can be determined according to national gradation requirements. In some embodiments, the median value of the quality requirement can be determined first based on the upper and lower limits of the quality requirements for the gradation interval in the national gradation requirements; then, a random perturbation can be applied based on this median value to determine the target quality for that gradation interval.

[0063] In some embodiments, the volume of the intermediate polyhedron can be calculated, and the volume of the intermediate polyhedron can be transformed to obtain the mass of the intermediate polyhedron.

[0064] In one example, the corresponding ballast gradation curve after randomly generating the ballast model is as follows: Figure 2 As shown.

[0065] It is understood that in this embodiment, the dual constraints of sieve aperture passability judgment and mass accumulation control can accurately control the mass and size distribution of ballast particles in each gradation interval during the generation stage, so that the mass distribution of the final generated ballast particle group falls within the gradation curve range required by the specification, without the need for further manual screening, which can improve the generation efficiency of ballast model.

[0066] Based on any of the above embodiments, the intermediate polyhedron can pass through the sieve holes, including: The sieve aperture size range is determined based on the gradation interval corresponding to the size range; If the area of ​​the target facet of the intermediate polyhedron is within the range of the sieve aperture size, then the intermediate polyhedron can pass through the sieve aperture.

[0067] Among them, the sieve aperture size range refers to the area range of the sieve apertures within the gradation interval corresponding to the size range.

[0068] The target facet refers to a representative facet on the intermediate polyhedron used to determine the passability of the sieve apertures.

[0069] For example, the sieve aperture can be a square aperture. Based on this, the area range of the sieve aperture can be specifically represented as the range of the squares of the upper and lower limits of the sieve aperture side length. The intermediate polyhedron that can pass through the sieve aperture can be determined by the following formula: (min_sieve)^2≤area_wh≤(max_sieve)^2) Where min_sieve is the lower limit of the sieve hole side length, max_sieve is the upper limit of the sieve hole side length, and area_wh is the area of ​​the target facet of the middle polyhedron.

[0070] It should be noted that there are multiple ways to select the target patch, such as selecting the largest cross section in the width and height direction, or selecting the main projection surface of the polyhedron, etc. This embodiment does not limit this.

[0071] Understandably, compared to the traditional method of judging sieve passability by simply comparing the length of the longest side, this embodiment, based on the median of the corresponding gradation interval and randomly perturbed to generate the length, width, and height of the intermediate polyhedral particles, checks whether the particles can pass through the sieve holes by measuring the area of ​​the target facet. This can accurately simulate the statistical probability of the intermediate polyhedral particles passing through the sieve holes after multiple flips during the sieving process, thus improving the precision of the sieve passability judgment.

[0072] Based on any of the above embodiments, it further includes: Calculate the target index of the intermediate polyhedron; the target index includes at least one of the needle-like index and the sheet-like index; If the target index is within the set index range, then the intermediate polyhedron is determined to be the target polyhedron; wherein, the ratio of the constraint mass to the gradation interval mass is within the set ratio range, and the constraint mass is obtained by accumulating the mass of the target polyhedron.

[0073] The target index is used to characterize the degree of excessive elongation of particles in the target direction. Specifically, the needle-like index is used to characterize the degree of excessive elongation of particles in the length direction, and the plate-like index is used to characterize the degree of excessive flattening of particles in the height direction.

[0074] Setting an index range refers to the range of values ​​for the needle-like index corresponding to extremely long and thin particles, and / or the range of values ​​for the flaky index corresponding to extremely flat particles, etc.

[0075] Constraint mass refers to the cumulative mass value of an intermediate polyhedron that meets the target index requirements such as needle-like and / or sheet-like indices.

[0076] For example, the range of values ​​for the needle-like index corresponding to extremely long and thin particles can be that the particle length exceeds a set multiple of the average diameter of the current gradation interval, such as 1.8 times. The range of values ​​for the flaky index corresponding to extremely flat particles can be that the particle height is less than a set multiple of the average diameter of the current gradation interval, such as 0.6 times.

[0077] Taking the target index, which includes needle-like index and flaky index, as an example, if the total mass of needle-like particles judged by the sieve aperture does not exceed 20% of the mass of the current gradation interval, it can be used as a random convex hull.

[0078] The current gradation interval refers to the gradation interval corresponding to the size range.

[0079] It should be noted that the specific values ​​of the set multiplier and the set ratio threshold can be adjusted according to the specific project requirements, and this embodiment does not limit them.

[0080] Understandably, compared to selecting the final ballast particle model from particles judged by their passability through the sieve based solely on geometric appearance or image measurement, this embodiment dynamically monitors target indices such as needle-like index and / or flaky index to dynamically constrain the cumulative mass of intermediate polyhedral particles that meet the target index requirements. This allows the proportion of intermediate polyhedral particles that meet the target index requirements in the overall particle group to be strictly limited, and further screening is performed using screening criteria that are closer to reality. This can improve the consistency between the generated ballast particle group and the real ballast in terms of statistical significance and mechanical behavior while avoiding substantial impact on the simulation results of the large-scale discrete element method.

[0081] At the same time, it can automatically skip the accumulation of subsequent particles of the same type when the mass proportion of a certain type of particle reaches the upper limit, avoiding the negative impact of subsequent post-processing such as manual screening on improving the generation efficiency of ballast models.

[0082] Based on any of the above embodiments, the size range of the ballast is determined according to its length, width, and height; The step of performing concave processing on at least two faces of the random convex hull at different scales includes: The average size is determined based on the length, the width, and the height; The average size is subjected to a first weighting process to obtain a small indentation parameter, and the average size is subjected to a second weighting process to obtain a large indentation parameter; Based on the small depression parameter and the large depression parameter, at least two facets of the random convex hull are subjected to depression processing of different scales.

[0083] The average dimension is a reference for calculating the indentation parameters, based on the length, width, and height. For example, the average dimension can be the arithmetic mean of the length, width, and height.

[0084] The first and second weighting processes refer to applying different weighting coefficients to the average size to obtain the corresponding small-scale depression parameters and large-scale depression parameters. The small-scale depression parameter is used to control the geometric dimensions of small-scale depressions, while the large-scale depression parameter is used to control the geometric dimensions of large-scale depressions.

[0085] It should be noted that the weighting coefficients of the first weighting process and the second weighting process can be adjusted based on engineering experience and the actual surface morphology of the ballast. This embodiment does not impose any restrictions on this.

[0086] Understandably, by processing the average size differently to obtain small and large indentation parameters, it is possible to simulate the coarse fracture surface and fine wear texture of ballast at different scales, thereby improving the consistency between the superimposed indentation effect and the multi-layered surface characteristics of real ballast.

[0087] Based on any of the above embodiments, the small depression parameters include the small depression radius and the small depression depth; The random convex hull surface is recessed according to the small recess parameter, including: The first target patch is randomly determined; Based on the fact that the depth of the small indentation decreases with the square of the distance, the vertices within the radius of the small indentation at the center of the first target facet are moved to perform indentation processing on the facet of the random convex hull.

[0088] Here, the small depression radius refers to the radius of a small-scale depression, and the small depression depth refers to the depth of a small-scale depression.

[0089] For example, the degree of decay with the square of the distance can be quantified by the following formula: ; in, It is the distance between the vertex to be moved and the center of the first target patch. It is the radius of the small-scale depression.

[0090] It should be noted that a weight that decays with the square of the distance can be determined, and vertices within the radius of the small depression at the center of the first target patch can be moved based on this weight and the small depression depth; alternatively, a bias that decays with the square of the distance can be determined, and vertices within the radius of the small depression at the center of the first target patch can be moved based on this bias and the small depression depth, etc. This embodiment does not limit this.

[0091] It should be noted that the selection of the first target patch can be done by uniform random selection or selection according to specific rules; the vertex movement direction can be the normal direction of the patch, or random perturbation can be superimposed on the normal direction of the patch, and this embodiment does not limit this.

[0092] Understandably, based on the fact that the depth of the small depression decreases with the square of the distance, moving the vertices within the radius of the small depression at the center of the first target patch can make the displacement of the pit edge transition smoothly, avoiding problems such as stepped cliffs or sharp abrupt changes, thereby simulating the natural transition effect of fine wear texture and enhancing the realism of the ballast particle model surface.

[0093] Based on any of the above embodiments, the large depression parameters include the large depression radius and the large depression depth; The random convex hull surface is concave according to the large concavity parameter, including: A second target patch is randomly determined; wherein, the distance between the centers of adjacent second target patches is not less than the interference distance, and the interference distance is obtained by a third weighting process based on the average size; Based on the fact that the depth of the large depression decreases with the cube of the distance, the vertices within the radius of the large depression at the center of the first target facet are moved to perform a depression process on the facet of the random convex hull.

[0094] Among them, the large depression radius refers to the radius of a large-scale depression, and the large depression depth refers to the depth of a large-scale depression.

[0095] For example, the degree of decay with distance cube can be quantified by the following formula: ; in, It is the distance between the vertex to be moved and the center of the second target patch. It is the radius of the large-scale depression.

[0096] It should be noted that a weight that decays with the cube of the distance can be determined, and vertices within the radius of the small depression at the center of the first target patch can be moved based on this weight and the small depression depth; alternatively, a bias that decays with the cube of the distance can be determined, and vertices within the radius of the small depression at the center of the first target patch can be moved based on this bias and the small depression depth, etc. This embodiment does not limit this.

[0097] The difference between the first and second target patches lies in the scale of the indentation treatment applied. Understandably, third, fourth, and other target patches may also exist.

[0098] The selection method for the second target patch is basically the same as that for the first target patch. The difference is that after obtaining the second candidate target patch, it is determined whether the distance between the centers of adjacent second candidate target patches is less than the interference distance. If it is less than the interference distance, a second candidate target patch is selected again. If it is not less than the interference distance, it is determined as the second target patch.

[0099] There are many ways to obtain the interference distance by performing a third weighting process based on the average size, such as by using preset weighting coefficients or preset mapping functions. This embodiment does not limit this method.

[0100] Understandably, based on the fact that the depth of a large depression decreases with the cube of the distance, moving the vertices within the radius of the large depression at the center of the first target patch can make the displacement of the central region of the depression greater and the edge of the depression steeper, thereby simulating a large-scale broken surface or crack on the ballast surface and enhancing the realism of the ballast particle model surface.

[0101] At the same time, by constraining the distance between adjacent second target patches by interference distance, it is possible to ensure that there is no overlapping interference between different large depressions, thus ensuring the uniformity and naturalness of the depression distribution.

[0102] See Figure 2 and Figure 3 This embodiment combines the square attenuation of small depressions with the cubic attenuation of large depressions to form a multi-scale roughness superposition effect, so that the final generated ballast particle model has the surface characteristics of real crushed stone at multiple scales.

[0103] Based on any of the above embodiments, after generating the ballast model, the method further includes: At least one of the model file and quality information of the ballast model is stored in a gradation range directory corresponding to the particle size distribution standard, based on the size range of the ballast model.

[0104] Depending on the actual needs, the model file export format can be at least one of the common 3D model formats such as STL, OBJ, and PLY; the quality information storage format can be at least one of the following: text file, spreadsheet, or database record; the naming rules for the gradation interval directory can be set according to the sieve aperture size interval, but this embodiment does not limit this.

[0105] For example, when entering a new gradation range or sieve aperture range, first create a corresponding folder, such as 50.0-63.0_mm; then for each ballast model that meets the conditions, call the export_to_stl function to export the ballast model as an STL file, and at the same time generate a .txt file with the same name to record the quality information of the ballast model; finally, store the STL file and the .txt file with the same name in the aforementioned corresponding folder.

[0106] Understandably, automatically classifying and storing ballast models and their quality information according to gradation intervals enables each sieve aperture interval's corresponding folder to contain all ballast particle models and their quality data that fall within that interval. This achieves a structured data management approach for catalogs, models, and quality labels, improving the usability and traceability of ballast models. It provides a foundation for users to quickly locate and select the required ballast model based on specific simulation needs, while also facilitating quality review and standardized verification of the generated ballast model results.

[0107] The above steps will be further explained below with reference to a specific embodiment.

[0108] The particle size distribution standard for premium grade crushed stone ballast can be determined first as follows: Then, each gradation interval of the particle size distribution standard can be determined as the corresponding ballast size range, and the length, width and height dimensions of different ballast particles can be generated cyclically by introducing appropriate random perturbations based on the midpoint of the interval.

[0109] The number of vertices of ballast particles, num_vertices, can be determined using the following formula: num_vertices=40+randi(70) Next, a random point set is generated within the stated size range; the intermediate polyhedron is then obtained based on the random point set. Specifically, we can first determine the coordinates of the vertices of each ballast particle in the range [0, length], [0, width], and [0, height] by adding a small Gaussian perturbation based on the length, width, and height of each particle, thus obtaining a random point set. The coordinates of each vertex are... A small Gaussian perturbation can be (+0.1*randn(...)).

[0110] Next, the extreme points corresponding to the minimum and maximum coordinates in the x, y, and z directions are selected from the random point set to obtain the extreme point set; four non-coplanar points are selected from the extreme point set to form an initial tetrahedron; the four triangular faces of the initial tetrahedron are used as initial faces. For example, it is possible to By taking the minimum / maximum coordinates of the direction, a maximum of 6 extreme point sets can be obtained. A triangle with four points can be constructed using the following formula. The plane equation: Where a is the x-component of the normal vector of the plane containing the triangle, b is the y-component of the normal vector of the plane containing the triangle, c is the z-component of the normal vector of the plane containing the triangle, and d is the distance correlation coefficient.

[0111] Calculate the fourth point Substitute the left end ,like They are not coplanar; based on this, from the set Select four non-coplanar points. These four points form the initial tetrahedron, and they determine the four initial faces. Each face is a triangle.

[0112] For each of the initial faces, calculate the directed distance of each point in the remaining point set relative to that face, and assign the points with directed distances greater than zero to the corresponding outer point set of that face; For example, for the remaining point set After assigning the points to the outer edges of each face according to their visibility, the vertices are... any side Its plane equation is: in, It is the x-component of the normal vector of the plane containing the k-th triangle. It is the y-component of the normal vector of the plane containing the k-th triangle. It is the z-component of the normal vector of the plane containing the k-th triangle. It is the distance correlation coefficient of the k-th triangle.

[0113] Each point to be assigned can be calculated using the following formula. Directed distance: like ,but In the face On the outside of the face, place it into the outer point set of the face. .

[0114] For faces with non-empty outer point sets, select the point farthest from the face, and form three new triangular faces with this point and the three edges of the face, respectively. Remove the original face and all faces visible to the new points. Reassign the points from the outer point sets of the removed faces to the new faces, and iterate the above steps until the outer point sets of all faces are empty. The directed distance is calculated as follows: determine the normal vector based on the three vertices of the face, substitute the point to be assigned into the plane equation of the face to obtain the signed distance value, and then perform normalization. The farthest point is selected by choosing the point in the outer point set that maximizes the calculated value of the plane equation. For example, for each non-empty face of the outer point set Execute: Select the farthest point and find the point with the maximum distance. : Will With the original face The three sides Each of them forms three new triangles; the original face And all of them Visible surfaces (i.e., those with which) All faces are removed from the convex hull; the point set is redistributed for each newly generated face. Filter pairs from the set of points of the removed faces For the points still on the outer edge, construct a new set of outer edges. This process is repeated until all points on the outer edges of the active surfaces are empty.

[0115] When no face has a non-empty set of outer points, all the remaining triangles form the final convex hull, and the output can be in the following form: Vertex list ; Patch Index Each face is represented by three vertex indices; After generating each random convex hull, at least two faces of each random convex hull can be concave at different scales to generate the corresponding ballast model.

[0116] Understandably, compared to simple, regular spheres or ordinary polyhedra, the above-mentioned concave treatment in this embodiment can make the surface of each ballast particle model present pits of varying sizes and depths, significantly enhancing the roughness and irregularity of the ballast particle model, and better restoring the fracture characteristics of real ballast.

[0117] The ballast model generation apparatus provided by the present invention will be described below. The ballast model generation apparatus described below and the ballast model generation method described above can be referred to in correspondence.

[0118] Figure 5 This is a schematic diagram of the structure of the ballast model generation device provided by the present invention, as shown below. Figure 5 As shown, the device includes: Particle size distribution standard determination module 510 is used to determine the particle size distribution standard of ballast. Size range determination module 520 is used to determine the size range of the ballast based on the particle size distribution standard; Random convex hull generation module 530 is used to generate random convex hulls within the said size range; The ballast model generation module is used to perform concave processing 540 on at least two facets of the random convex hull at different scales to generate a ballast model.

[0119] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6 As shown, the electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communications interface 620, and the memory 630 communicate with each other via the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute the following methods: Determine the particle size distribution standard for ballast; obtain the size range of ballast based on the particle size distribution standard; generate a random convex hull within the size range; perform concave processing on at least two facets of the random convex hull at different scales to generate a ballast model.

[0120] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to related technologies, or a portion 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.

[0121] This invention discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer can perform the methods provided in the above-described method embodiments, such as including: Determine the particle size distribution standard for ballast; obtain the size range of ballast based on the particle size distribution standard; generate a random convex hull within the size range; perform concave processing on at least two facets of the random convex hull at different scales to generate a ballast model.

[0122] On the other hand, embodiments of the present invention also provide a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is implemented to perform the transmission methods provided in the above embodiments, including, for example: Determine the particle size distribution standard for ballast; obtain the size range of ballast based on the particle size distribution standard; generate a random convex hull within the size range; perform concave processing on at least two facets of the random convex hull at different scales to generate a ballast model.

[0123] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0124] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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 generating a ballast model, characterized in that, include: Determine the particle size distribution standard for ballast; The size range of the ballast is obtained based on the particle size distribution standard; Generate a random convex hull within the specified size range; At least two faces of the random convex hull are concave at different scales to generate a ballast model.

2. The method for generating a ballast model according to claim 1, characterized in that, The determination of the ballast size range based on the particle size distribution standard includes: At least two gradation intervals are obtained based on the particle size distribution standard; The size range of the ballast is obtained by introducing random disturbances into each of the gradation intervals.

3. The method for generating a ballast model according to claim 1, characterized in that, Generating a random convex hull within the specified size range includes: An intermediate polyhedron is generated within the specified size range; If the intermediate polyhedron can pass through the sieve holes, the random convex hull is obtained based on the intermediate polyhedron; wherein, the gradation interval quality is within the set quality range of the gradation interval, and the gradation interval quality is obtained by accumulating the quality of the random convex hull.

4. The method for generating a ballast model according to claim 3, characterized in that, The intermediate polyhedron can pass through the sieve holes, including: The sieve aperture size range is determined based on the gradation interval corresponding to the size range; If the area of ​​the target facet of the intermediate polyhedron is within the range of the sieve aperture size, then the intermediate polyhedron can pass through the sieve aperture.

5. The method for generating a ballast model according to claim 4, characterized in that, Also includes: Calculate the target index of the intermediate polyhedron; the target index includes at least one of the needle-like index and the sheet-like index; If the target index is within the set index range, then the intermediate polyhedron is determined to be the target polyhedron; wherein, the ratio of the constraint mass to the gradation interval mass is within the set ratio range, and the constraint mass is obtained by accumulating the mass of the target polyhedron.

6. The method for generating a ballast model according to claim 1, characterized in that, The dimensions of the ballast are determined based on its length, width, and height; The step of performing concave processing on at least two faces of the random convex hull at different scales includes: The average size is determined based on the length, the width, and the height; The average size is subjected to a first weighting process to obtain a small indentation parameter, and the average size is subjected to a second weighting process to obtain a large indentation parameter; Based on the small depression parameter and the large depression parameter, at least two facets of the random convex hull are subjected to depression processing of different scales.

7. The method for generating a ballast model according to claim 6, characterized in that, The parameters of the small indentation include the radius of the small indentation and the depth of the small indentation; The random convex hull surface is recessed according to the small recess parameter, including: The first target patch is randomly determined; Based on the fact that the depth of the small indentation decreases with the square of the distance, the vertices within the radius of the small indentation at the center of the first target facet are moved to perform indentation processing on the facet of the random convex hull.

8. The method for generating a ballast model according to claim 6, characterized in that, The large depression parameters include the large depression radius and the large depression depth; The random convex hull surface is concave according to the large concavity parameter, including: A second target patch is randomly determined; wherein, the distance between the centers of adjacent second target patches is not less than the interference distance, and the interference distance is obtained by a third weighting process based on the average size; Based on the fact that the depth of the large depression decreases with the cube of the distance, the vertices within the radius of the large depression at the center of the first target facet are moved to perform a depression process on the facet of the random convex hull.

9. The method for generating a ballast model according to any one of claims 1-8, characterized in that, After generating the ballast model, the process also includes: At least one of the model file and quality information of the ballast model is stored in a gradation range directory corresponding to the particle size distribution standard, based on the size range of the ballast model.

10. A ballast model generating device, characterized in that, include: The particle size distribution standard determination module is used to determine the particle size distribution standard of ballast. A size range determination module is used to determine the size range of the ballast based on the particle size distribution standard. A random convex hull generation module is used to generate a random convex hull within the stated size range; The ballast model generation module is used to perform concave processing on at least two facets of the random convex hull at different scales to generate a ballast model.

Images