CN115618702B - Generation Method of 2D Numerical Simulation Model of Ultra-high Volume Fraction Mélange Using Overlapping Contact Cutting Algorithm - Google Patents

Abstract

The invention discloses a method for generating a two-dimensional ultra-high volume fraction mixed rock numerical simulation model by adopting an overlapped contact cutting algorithm, which comprises the steps of randomly generating a polygonal aggregate frame according to preconditions such as a given grain grading curve; recording coordinate information of a polygonal aggregate frame, introducing discrete meta-software, placing the aggregate frame into a boundary area wall, and realizing the conversion of aggregate particles into rigid clusters; performing DEM simulation to enable aggregates to collide freely until balance, and recording position information distributed after cluster collision in a geometric file; and then judging the contact property of aggregate entities in the geometric file by calling an aggregate contact judging algorithm developed by Python language, and cutting aggregates with contact and overlapping quantities to finally obtain the numerical simulation model of the high-volume-fraction mixed rock. The invention can generate the mixed rock numerical simulation model with extremely high volume fraction, avoid overlapping among aggregates, and better simulate the contact surface among aggregate particles.

Description

Generation Method of 2D Numerical Simulation Model of Ultra-high Volume Fraction Mélange Using Overlapping Contact Cutting Algorithm

technical field

The invention belongs to the technical field of research on numerical simulation parameters of heterogeneous rock-soil materials mixed rocks, and in particular relates to a method for generating a two-dimensional ultra-high volume fraction mixed rock numerical simulation model using an overlapping contact cutting algorithm.

Background technique

Due to the limitations of experimental methods and the rapid rise and development of numerical simulation methods, the use of numerical simulation methods to study the mechanical parameters and mechanical properties of mélange materials can better overcome the problems of sample integrity and in-situ experiments and laboratory experiments. The disturbance of the sample, the size of the sample, and the errors caused by the experimental operation.

There is an important problem in numerical simulation research—the establishment of experimental model, that is, to generate numerical model samples with aggregate gradation and high aggregate volume fraction that meet the requirements of numerical simulation. At present, there are basically no researchers who have generated mélange models with an aggregate volume fraction greater than 90%; however, in laboratory tests, the aggregate volume fraction of mélange samples has reached 90%, such as Afifipour et al. in Mechanical behavior of bimrocks having In the article "high rock block proportion", the aggregate volume fraction of melange is increased to 90%. Therefore, it is necessary to study the numerical simulation model of ultra-high-volume mélange to overcome problems such as errors in in-situ experiments and laboratory experiments.

In terms of model establishment, there are mainly two methods for generating mesoscopic numerical simulation models, one is the image processing method; the other is the random block placement method. The image processing method is to scan and image the cross-section of the melange sample, so as to distinguish the aggregate and the matrix, and establish a model; but this method cannot study the randomness of the material distribution inside the melange material, and has certain limitations sex. The random block placement method generates random blocks through an algorithm, and uses the placement algorithm to place blocks within the specified model boundary area, but the contact must be considered when placing, so the aggregate volume fraction of the generated melange model is difficult to meet the requirements, and even more difficult Generates melange models with very high volume fractions.

Contents of the invention

The purpose of the present invention is to provide a method for generating a two-dimensional ultra-high volume fraction melange numerical simulation model using an overlapping contact cutting algorithm, so as to solve the problem that the ultra-high volume fraction melange model and the simulated aggregate interface cannot be truly generated in the prior art Technical issues with surface contact.

In order to solve the above technical problems, the present invention adopts the following technical solutions to realize:

A method for generating a two-dimensional ultra-high volume fraction melange numerical simulation model using an overlapping contact cutting algorithm, comprising the following steps:

Step 1: Generate all polygonal aggregate frame geometric models based on information such as particle size distribution curve and aggregate volume fraction.

Step 2: Use discrete element software to generate rigid cluster clumps of the same size and shape according to the polygonal aggregate frame, and set the model boundary as the wall boundary; ensure that the aggregate particles can be within the model boundary, and at the same time simulate the aggregate in the boundary distribution on

Step 3: After generating rigid clusters of aggregates and setting boundaries, use discrete elements to simulate the collision of rigid clusters to redistribute the positions of aggregates; obtain the coordinate information of the outline of rigid clusters of aggregates after redistribution after the collision of aggregate particles reaches equilibrium, Obtain the melange geometric model with random distribution of aggregate particles and save it as a geometry file; this step can simulate the random distribution of aggregate particles in melange in nature, allowing aggregate particles to be distributed at any position and at any angle, close to the real situation;

Step 4: Call the contact overlap judgment algorithm to process the melange model geometry file; obtain the polygonal entities of all aggregate particles in the entire model, through the above algorithm, loop through all aggregate entities, and find contact overlap with the surrounding polygonal aggregate particles amount of aggregate, and to identify the type of contact overlap.

Step 5: According to the type of aggregate contact overlap, different contact overlap cutting algorithms are used to cut the overlapped aggregates, and the corresponding aggregate information is updated to the list of polygonal aggregate entities after each cutting.

Step 6: Repeat step 5 for the updated aggregate entity list until all polygonal aggregates within the boundaries that generate contact overlap are cut and updated. For aggregate particles that touch and overlap after cutting, if the aggregate entity is not updated, once a certain aggregate contacts and overlaps with multiple aggregates, when the geometry file is re-exported later, the first cut overlapping part will be drawn repeatedly, resulting in cutting Failed; therefore, after step 5 each time, the aggregate entity list needs to be updated until there is no contact overlap between any two aggregates.

Step 7: Use the corresponding geometry export algorithm to rewrite the cut polygonal aggregate into a geometry file and export it; and accumulate the area of the aggregate particle entity in the exported geometry file to obtain the final 2D melange model The total area S _a of the aggregate particles in the medium is calculated as a ratio of the area S _b of the boundary area of the model to obtain the volume fraction of the aggregate after cutting.

Further optimization, in step 1, the process of generating all polygonal aggregate frames is as follows:

Step 1.1: In the polar coordinate system, take the origin as the base point, obtain a series of random radius and angle values through random functions, obtain a series of random points according to the random radius and angle, and connect these points sequentially to obtain polygonal aggregates;

Step 1.2: Obtain the total area of all aggregates within the model boundary by accumulating; divide according to the aggregate particle size interval of the particle gradation curve, and calculate according to the ratio of the area of the aggregate in the corresponding particle size interval to the total area of all aggregates Control the proportion of aggregate in the model in the corresponding particle size range;

Step 1.3: The total area of all aggregates is obtained by summing up, and the volume fraction of the melange model is controlled by the ratio of the total area of aggregates to the area of the boundary area of the model.

Further optimization, the step 2 also includes the following steps: using the Minkowski algorithm to expand outwards with the normal direction of each side of the polygonal aggregate model, so that the outlines of all aggregates are extended outwards; using discrete elements The collision of clusters is simulated, and after the collision of aggregate particles reaches equilibrium, the outer contour of the aggregate is restored to the state before expansion. By extending the boundary of the aggregate frame outward, it is equivalent to wrapping a shell around the aggregate, so that part of the overlap generated during the collision of the aggregate is in the shell part. When the outer contour of the aggregate is restored to the state before the expansion, The contact overlap between some aggregates will disappear, that is, the overlap between aggregates is reduced, and the contact and overlap of aggregate particles can be controlled initially by controlling the distance of expansion.

Further optimization, in the step 2, the method for generating rigid clusters is as follows: generate a template for rigid clusters with the polygonal frame of each aggregate particle, use this template as the basis for generating rigid clusters with the same shape as the polygonal aggregate, and then pass Use the Bubble Pack algorithm to automatically fill the rigid cluster template with Pebble particles to obtain rigid clusters. There is no relative deformation between the Pebble particles inside the rigid cluster (clump), so it will not be damaged, which ensures that the aggregate particles will not be deformed or broken during the collision, and the original aggregate shape will be guaranteed.

Further optimization, in the step 3, the collision characteristics of rigid clusters are as follows: in the discrete element software, the collision between rigid clusters is simulated by the discrete element method, the boundary of the model is set as a rigid wall, and stiffness is given, and aggregate The rigid cluster also endows the contact model and stiffness. Since both the wall and the aggregate have stiffness, they will collide and bounce apart when they overlap; Collision between them, so as to realize the redistribution of aggregate particles, and allow the presence of aggregate particles on the boundary.

For further optimization, in order to ensure that the aggregate particles do not fly out of the boundary during the collision, the stiffness of the wall is much greater than that of the aggregate rigid cluster; the contact between the aggregate and the wall will be more difficult than the contact between the aggregate rigid cluster; when the volume When the score is set larger, due to the characteristics of the overlapping amount allowed by the discrete element method, there will be a certain amount of overlap between the aggregate rigid clusters, which is inconsistent with the actual situation.

In order to ensure the volume fraction required by the model, for the aggregate particles located on one side of the boundary wall, the part beyond the side boundary is placed on the other side boundary that is symmetrical to it through the periodic boundary; Aggregate particles are divided into four parts, which are respectively placed inside the four corner points corresponding to the model boundary through periodic boundaries.

Further optimization, in the step 4, the contact overlap judgment algorithm is used to judge the aggregate contact overlap type, as follows: the contact overlap judgment algorithm is used to judge the aggregate contact overlap type, as follows: read the geometry file, and obtain each aggregate polygon The coordinate point information; then call the polygon contact overlap judgment algorithm to judge the contact overlap amount of the aggregate polygon.

The contact overlap of aggregates includes the following seven types:

1) There is point-to-point contact between adjacent polygonal aggregates;

2) There is point-to-line contact between adjacent polygonal aggregates;

3) There is line-to-line contact between adjacent polygonal aggregates;

4) There is overlap between adjacent polygonal aggregates, and the overlapping part is a polygon, and the overlapping part is inside the model boundary and has no contact with the boundary;

5) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons, and the overlapping part is inside the model boundary and has no contact with the boundary;

6) There is an amount of overlap between adjacent polygonal aggregates, and the overlapping part is a polygon. At the same time, the overlapping part is in contact with the model boundary, and there is only one intersection point between the two polygonal aggregates.

7) There is an amount of overlap between adjacent polygonal aggregates, and the overlapping part is a polygon. At the same time, the overlapping part is in contact with the model boundary, and there are multiple intersection points between the two polygonal aggregates.

Further optimization, in the step 5, in the step 5, different cutting methods are implemented according to different contact overlapping types, as follows:

1) Point-to-point contact between adjacent polygonal aggregates, simulating the contact between sharp corners and sharp corners between aggregate particles in melange in reality, which is allowed to exist without cutting.

2) Point-to-line contact between adjacent polygonal aggregates, simulating the contact between sharp corners and surfaces between aggregate particles in melange in reality, which is allowed and does not need to be cut.

3) The line-to-line contact between adjacent polygonal aggregates simulates the surface-to-surface contact between aggregate particles in melange in reality, which is allowed and does not need to be cut.

4) There is overlap between adjacent polygonal aggregates, and the overlapping part is a polygon, and the overlapping part is located inside the model boundary and has no contact with the boundary.

The overlapping aggregates for this situation are cut as follows:

4.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint Polygonal aggregates, with overlapping parts removed.

4.2) Adopt the geometric figure intersection discrimination method to obtain the intersection of two polygonal aggregates with overlapping amounts; read the geometric file, read the vertex information of the corresponding geometric figure, and obtain the set of coordinates of each vertex of the polygon corresponding to the overlapping part, which is recorded as The first set; and the set of the vertex coordinates of the two aggregates that overlap, which is recorded as the second set; then the intersection of the first set and the second set is calculated, which is recorded as the third set; finally, from the first set Delete the vertex coordinates in the third set, and the remaining two vertices are the intersection points of the two aggregates. Since the intersection point is located on the edge of the polygonal aggregate, it cannot be found directly, so it needs to be searched by the above method.

4.3) Make a cutting line through these two intersection points, and cut the combined polygonal aggregate into two aggregates without overlapping along the cutting line.

5) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons. It is inside the model boundary and has no contact with the boundary. In this case, the overlapping aggregates are cut according to the following method:

5.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint Polygonal aggregates, with overlapping parts removed.

5.2) Using the intersection discrimination method of geometric figures, the intersection of two aggregates with overlapping amounts is obtained. The intersection is a plurality of polygons, and the coordinates of the intersection point cannot be obtained directly. It is necessary to further traverse the intersection to obtain the vertex coordinate set of each polygon.

5.3) Calculate the distance between any two vertex coordinate points in the intersection, find the two coordinate points with the largest distance as the intersection point of the aggregate, make a cutting line through these two intersection points, and combine polygonal aggregates along the cutting line Cut into two aggregates with no overlapping amounts.

6) There is an overlapping amount between adjacent polygonal aggregates, and the overlapping part is a polygon, and at the same time the overlapping part is in contact with the model boundary, the overlapping part is divided into two parts by the boundary of this side, one part is located near the side boundary, and the other part The periodical boundary is placed near the boundary on the other side that is symmetrical to it; there is only one intersection point between the two shaped aggregates in each part, and the overlapping aggregates for this case are cut as follows:

6.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint Polygonal aggregates, with overlapping parts removed.

6.2) Use the geometrical figure intersection discrimination method to obtain the intersection of two aggregates with overlapping amounts. Since the overlapping part is in contact with the boundary, there is only one intersection point in the overlapping part. Use a method similar to step 4.2) to find the intersection point; then The vertex coordinates of the intersection are traversed and judged, and any vertex on the boundary is found as another intersection point.

6.3) Make a cutting line through these two intersection points, and cut the combined polygonal aggregate into two aggregates without overlapping along the cutting line; the other part of the overlapping amount located near the symmetrical boundary is cut using the same method as above.

7) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons. At the same time, the overlapping part is in contact with the model boundary, then the overlapping part is divided into two parts by the boundary of this side, one part is located near the side boundary, and the other One part is placed near the border on the other side which is symmetrical to it through the periodic boundary; the two shaped aggregates in each part only have multiple intersections, and the overlapping aggregates for this case are cut as follows:

7.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint Polygonal aggregates, with overlapping parts removed.

7.2) Using the intersection discrimination method of geometric figures, the intersection of two aggregates with overlapping amounts is obtained. Since the overlapping part is in contact with the boundary, there are multiple intersection points in the overlapping part, and the coordinates of the intersection point cannot be obtained directly. It is necessary to further traverse the intersection to obtain A collection of vertex coordinates for each polygon.

7.3) Calculate the distance between any two vertex coordinate points in the intersection, find the two coordinate points with the largest distance as the intersection point of the aggregate, make a cutting line through these two intersection points, and combine polygonal aggregates along the cutting line Cut into two aggregates with no overlapping amount; the other overlapping amount located near the symmetrical boundary is cut in the same way as above.

Further optimization, in the step 7, the method for calculating the volume fraction of the cut melange model aggregate is specifically:

Step 7.1: read the geometric model file of the cut polygonal aggregate;

Step 7.2: Calculate the polygonal area of each aggregate particle after cutting, and add up the total area S _a of all aggregates after cutting;

Step 7.3: Calculate the ratio of the total area S _a of all aggregates after cutting to the area S _b of the boundary area of the model to obtain the volume fraction of aggregates after cutting. Since the model is a two-dimensional model, after being converted into a three-dimensional model, it becomes a sheet-like structure of equal thickness, so the volume content can be expressed by the area ratio.

Compared with the prior art, the present invention has the following beneficial effects:

The invention well solves the problems that the existing ultra-high volume fraction melange model is difficult to generate and the aggregates have contact overlap. Through the contact overlap cutting algorithm between aggregates, the requirements of surface-to-surface contact between aggregates and ultra-high volume fraction melange model generation are realized. This method converts the problem of the difficulty in generating the numerical model of ultra-high volume fraction melange and the contact overlap between aggregates into surface-to-surface contact between aggregates through a cutting algorithm. The generated model is more realistic and can improve the numerical simulation model. volume fraction and authenticity.

Description of drawings

Fig. 1 is the flow chart of the generation method of the two-dimensional ultra-high volume melange numerical simulation model that adopts overlapping contact cutting algorithm according to the present invention;

Fig. 2 is the simulation figure of aggregate frame in embodiment 1;

Fig. 3 is the simulated diagram of aggregate wrapping Minkowski shell in embodiment 1;

Fig. 4 is that aggregate is converted into rigid cluster in embodiment 1, and the simulation figure after carrying out discrete element collision;

Fig. 5 is after the collision in embodiment 1, the simulated figure after aggregate dispersion equilibrium;

Fig. 6 is part A in Fig. 5, a partial enlarged view of the point-to-point contact between the aggregates in the melange model;

Fig. 7 is part B in Fig. 5, a partial enlarged view of point-to-line contact between aggregates in the melange model;

Figure 8 is a partial enlarged view of part C in Figure 5, where the contact overlap between the aggregates in the melange model is a single polygon and inside the boundary;

Fig. 9 is a schematic diagram of the cutting steps in which the contact overlap between the aggregates in the melange model is a polygon and is inside the boundary;

Fig. 10 is part D in Fig. 5, the partial enlarged view of the contact overlapping part between the aggregates in the melange model is a plurality of polygons and inside the boundary;

Fig. 11 is a schematic diagram of the cutting steps in Example 1 for the contact overlap between the aggregates in the melange model as multiple polygons and inside the boundary;

Fig. 12 is part E in Fig. 5, a partial enlarged view of the contact overlapping part between the aggregates in the melange model on the boundary;

Fig. 13 is a schematic diagram of the cutting steps on the boundary of the contact overlapping part between the aggregates in the melange model in Example 1;

Fig. 14 is the final model diagram after the contact overlap judgment and cutting processing are performed on the melange model in embodiment 1;

Fig. 15 is set aggregate volume fraction as 95% mélange model in embodiment 2, the simulated figure after aggregate dispersion balance;

Fig. 16 is the final model figure after the contact overlap judgment and cutting processing are performed on the melange model in embodiment 2;

Fig. 17 is that the aggregate volume fraction is set to be 100% mélange model in embodiment 3, the simulated figure after aggregate dispersion balance;

Fig. 18 is the final model diagram after performing contact overlap judgment and cutting processing on the melange model in embodiment 3.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings; obviously, the embodiments described below are some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Example 1:

As shown in Figure 1, a method for generating a numerical simulation model of a two-dimensional ultra-high volume fraction melange using the overlapping contact cutting algorithm includes the following steps:

Step 1: Use Matlab software to generate a polygonal aggregate geometric model based on the grading curve and aggregate volume fraction information.

In this embodiment, the boundary size of the two-dimensional mixed rock model is set to be 10cm*20cm, and the volume fraction of aggregate is 90%. The particle size distribution of particle gradation is the following four intervals: [6.0, 9.0], [9.0, 10.5], [10.5, 11.2], [11.2, 12], the unit is mm. The volume percentages of the aggregate in the corresponding particle size range to the total aggregate are: 19.6%, 47.9%, 24.5%, 8.0%, as shown in Figure 2, the polygon is the geometric model of the aggregate. This method has no special requirements on the particle gradation of the aggregate, which can be set arbitrarily.

Step 2: Use the Minkowski algorithm to expand outward with the normal direction of each side of the polygonal aggregate model, so that the outline of all aggregates extends outward, which is equivalent to wrapping the aggregate with a shell, as shown in Figure 3 Show. Use discrete element software to generate rigid clusters (clumps) of the same size and shape with the polygonal framework of each aggregate particle to generate a template for rigid clusters, and use this template as the basis for generating rigid clusters with the same shape as polygonal aggregates, and then use Bubble The Pack algorithm automatically fills the rigid cluster template with Pebble particles to obtain a rigid cluster (clump), as shown in Figure 4, and sets the model boundary as a periodic rigid wall boundary, which can simulate the distribution of aggregates on the boundary.

Step 3: Use discrete elements to simulate the collision of rigid clusters to redistribute the aggregate positions, as shown in Figure 5. Then restore the external profile of the aggregate to the state before expansion. Set the stiffness of the boundary wall of the model to be much greater than the rigidity of the aggregate rigid cluster; ensure that the aggregate particles will not fly out of the boundary during the collision; The boundary is placed on the other side of the boundary that is symmetrical to it; the aggregate particles located at the corners of the model boundary are divided into four parts, and are placed inside the four corresponding corners of the model boundary through periodic boundaries. Then obtain the coordinate information of the aggregate rigid cluster outline after redistribution, obtain the melange geometric model with random distribution of aggregate particles, and save it as a dxf file.

Step 4: Use the Python contact overlap judgment algorithm to process the dxf file of the melange geometric model; obtain the polygonal entities of all aggregate particles in the entire model, and use the above algorithm to loop through all aggregate entities to find out the existence of polygonal aggregate particles with the surrounding polygons. Aggregate with contact overlap amount, and identify contact overlap type.

Step 5: According to the type of aggregate contact overlap, different contact overlap cutting algorithms are used to cut the overlapped aggregates, and the corresponding aggregate information is updated to the list of polygonal aggregate entities after each cutting.

Different cutting methods are implemented according to different contact overlapping types, as follows:

1) Point-to-point contact between adjacent polygonal aggregates, simulating the contact between sharp corners and sharp corners between aggregate particles in melange in reality, which is allowed to exist without cutting, as shown in Figure 6.

2) Point-to-line contact between adjacent polygonal aggregates, simulating the contact between sharp corners and surfaces between aggregate particles in melange in reality, which is allowed to exist without cutting, as shown in Figure 7.

3) The line-to-line contact between adjacent polygonal aggregates simulates the surface-to-surface contact between aggregate particles in melange in reality, which is allowed and does not need to be cut.

4) There is an overlapping amount between adjacent polygonal aggregates, and the overlapping part is a polygon. The overlapping part is located inside the model boundary and has no contact with the boundary. As shown in Figure 8, the overlapping aggregates for this situation are performed according to the following method cutting:

Using the efficient combination method of the shapely library in Python, the overlapping aggregates are combined to form a combined polygonal aggregate, and the intermediate contact overlapping part is deleted; at the same time, through the intersection method in the shapely library, the two overlapped The intersection of all polygonal aggregates; through the ponits method of the entity in the dxfgrabber library, obtain the set of coordinates of each vertex of the polygon corresponding to the overlapping part, which is recorded as the first set; and the set of coordinates of each vertex of the two aggregates that overlap , recorded as the second set; then calculate the intersection of the first set and the second set, and record it as the third set; finally delete the vertex coordinates in the third set from the first set, and the remaining two vertices are two The intersection point of the aggregate; then use these two intersection points as the cutting line, and use the split method in the shapely library to cut the joint polygon aggregate into two aggregates without overlapping, as shown in Figure 9.

5) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons. It is inside the model boundary and has no contact with the boundary. As shown in Figure 10, the overlapping aggregates in this case are cut according to the following method :

Use the efficient combination method of the shapely library in Python to combine the overlapping aggregates to form a combined polygonal aggregate, and the intermediate contact overlapping part is deleted; at the same time, through the intersection method in the shapely library, return this aggregate and The intersection of another aggregate, but at this time the intersection is multiple polygons, and the intersection vertex coordinates cannot be obtained directly. It is necessary to further traverse the intersection to obtain the vertex coordinates of each single polygon; then, call the scipy library and numpy library to calculate each intersection in the intersection. The distance between two coordinate points, find the two coordinate points with the largest distance as the intersection point of the aggregate, use these two intersection points as the cutting line, and use the split method in the shapely library to cut the joint polygon aggregate into two without Overlapping amounts of aggregates, as shown in Figure 11.

6) There is an overlapping amount between adjacent polygonal aggregates, and the overlapping part is a polygon, and at the same time the overlapping part is in contact with the model boundary, the overlapping part is divided into two parts by the boundary of this side, one part is located near the side boundary, and the other part The periodical boundary is placed near the boundary on the other side that is symmetrical to it; there is only one intersection point between the two shaped aggregates in each part, as shown in Figure 12, and the overlapping aggregates for this situation are cut as follows:

Use the efficient combination method of the shapely library in Python to combine aggregates with contact overlap to form a combined polygonal aggregate, and delete the overlapping part; use the intersection method in the shapely library to obtain two aggregates with overlapping amounts The intersection of , because the overlapping part is in contact with the boundary, there is only one intersection point in the overlapping part, through the ponits method of the entity in the dxfgrabber library, use a method similar to step 4.2) to find the intersection point; then traverse and judge the vertex coordinates of the intersection, Find any vertex on the boundary as another intersection point; make a cutting line through these two intersection points, and use the split method in the shapely library to cut the joint polygon aggregate into two aggregates without overlapping, as shown in Figure 13 Show. Another part of the overlapping volume located near the symmetric boundary is cut in the same way as above.

7) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons. At the same time, the overlapping part is in contact with the model boundary, then the overlapping part is divided into two parts by the boundary of this side, one part is located near the side boundary, and the other One part is placed near the border on the other side which is symmetrical to it through the periodic boundary; the two shaped aggregates in each part only have multiple intersections, and the overlapping aggregates for this case are cut as follows:

Use the efficient combination method of the shapely library in Python to combine aggregates with contact overlap to form a combined polygonal aggregate, and delete the overlapping part; use the intersection method in the shapely library to obtain two aggregates with overlapping amounts Since the overlapping part is in contact with the boundary, there are multiple intersection points in the overlapping part, and the coordinates of the intersection point cannot be obtained directly. It is necessary to further traverse the intersection, and obtain the vertex coordinate set of each polygon through the ponits method of the entity in the dxfgrabber library; call The scipy library and the numpy library calculate the distance between any two vertex coordinate points in the intersection, find the two coordinate points with the largest distance as the intersection point of the aggregate, and make a cutting line through these two intersection points, using the shapely library The split method cuts the joint polygon aggregate into two aggregates with no overlapping volume; the other part of the overlapping volume located near the symmetrical boundary is cut using the same method as above.

Step 6: Repeat step 5 for the updated aggregate entity list until all polygonal aggregates within the boundaries that generate contact overlap are cut and updated. After being processed by the contact overlap cutting algorithm, two polygonal aggregates with contact overlap are cut into two polygonal aggregates with only one line coincident contact.

Step 7: Use the corresponding dxfgrabber library in Python to rewrite the cut polygonal aggregate as a dxf file and export it. The final geometric model is shown in Figure 14 (the difference between Figure 14 and Figure 5 is that some There is a contact overlap between adjacent aggregates, and the overlapping parts between adjacent aggregates in Figure 14 have been cut); then, call the corresponding shapely library of Python, use the area method of the polygon, and export The total area of aggregate particles in the dxf file is accumulated to obtain the total area S _a of aggregate particles in the final two-dimensional melange model, and the ratio calculation is performed with the area S _b of the boundary area of the model to obtain the aggregate area after cutting Volume fraction. Since the coordinates of the vertex corresponding to the deformation of the aggregate after cutting are known, the area of each aggregate is also known. The volume fraction of the aggregate in this example model after cutting is 89.17%, and the error between theory and practice is 0.83%, which is less than the set error value of 1%, so it meets the volume fraction set by the model.

Embodiment two:

Using the same method as in Example 1, generate a melange model in which the aggregate volume fraction is set to 95%, and the simulation diagram after the aggregate dispersion and balance is shown in Figure 15; after processing through the cutting algorithm, the aggregate volume fraction is obtained The geometric model of the mélange is 92.26%, as shown in Fig. 16.

Embodiment three:

Using the same method as in Example 1, generate a mélange model with the aggregate volume fraction set to 100%, and the simulation diagram after the aggregate dispersion and balance is shown in Figure 17; after the points are processed by the cutting algorithm, the aggregate volume fraction is obtained is 95.25% of the melange geometric model, as shown in Figure 18. As long as the set volume fraction increases, the method of the present invention can generate a geometric model with an aggregate volume fraction close to 100%.

The invention is not only applicable to the generation of the numerical simulation model of super-high volume fraction mixed rock, but also applicable to heterogeneous rock and soil materials such as concrete and earth-rock mixture.

The above is only an embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the description of the present invention and the contents of the accompanying drawings, or directly or indirectly used in other related technologies fields, are all included in the scope of patent protection of the present invention in the same way.

Claims (9)

1. The generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting the overlapping contact cutting algorithm is characterized in that, comprising the following steps: Step 1: Generate the two-dimensional framework geometric models of all polygonal aggregates according to the particle gradation curve and aggregate volume fraction information; Step 2: Use discrete element software to generate rigid clusters of the same size and shape according to the polygonal aggregate frame, and set the model boundary as the wall boundary to ensure that the aggregate particles can be within the model boundary and simulate the aggregate on the boundary Distribution; Step 3: Use discrete elements to simulate the collision of rigid clusters to redistribute the aggregate positions; obtain the coordinate information of the outline of the redistributed aggregate rigid clusters after the collision of aggregate particles reaches equilibrium, and obtain the melange geometry with random distribution of aggregate particles model, and save it as a geometry file; Step 4: Use the contact overlap judgment algorithm to process the melange model geometry file, obtain the polygonal entities of all aggregate particles in the entire model, loop through all aggregate entities, and find aggregates that have contact overlap with the surrounding polygonal aggregate particles, And distinguish the type of contact overlap; Step 5: According to the type of aggregate contact overlap, use different contact overlap cutting algorithms to cut the contact overlap aggregates, and update the corresponding aggregate information to the list of polygon aggregate entities after each cut; Step 6: Repeat step 5 for the updated aggregate entity list until all polygonal aggregates within the boundaries that generate contact overlap are cut and updated; Step 7: rewrite the polygonal aggregate after cutting into a geometry file and export it; and accumulate the area of the aggregate particle entity in the exported geometry file to obtain the total area S of the aggregate particle in the final two-dimensional hybrid rock model _a , and calculate the ratio with the model boundary area S _b to obtain the volume fraction of the aggregate after cutting. 2. the generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting overlapping contact cutting algorithm according to claim 1, it is characterized in that, in the described step 1, the process of generating all polygonal aggregate frames is as follows : Step 1.1: In the polar coordinate system, take the origin as the base point, obtain a series of random radius and angle values through random functions, obtain a series of random points according to the random radius and angle, and connect these points sequentially to obtain polygonal aggregates; Step 1.2: Obtain the total area of all aggregates within the model boundary by accumulating; divide according to the aggregate particle size interval of the particle gradation curve, and calculate according to the ratio of the area of the aggregate in the corresponding particle size interval to the total area of all aggregates Control the proportion of aggregate in the model in the corresponding particle size range; Step 1.3: The total area of all aggregates is obtained by summing up, and the volume fraction of the melange model is controlled by the ratio of the total area of aggregates to the area of the boundary area of the model. 3. the generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting overlapping contact cutting algorithm according to claim 2, it is characterized in that, in described step 2, also comprise the following steps: adopt Minkowski The algorithm uses the normal direction of each side of the polygonal aggregate model to expand outward, so that the outline of all aggregates extends outward; uses discrete elements to simulate the collision of clusters, and after the collision of aggregate particles reaches equilibrium, the aggregate The outer contour reverts to what it was before the extension. 4. the generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting overlapping contact cutting algorithm according to claim 3, it is characterized in that, in described step 2, the method for generating rigid cluster is specifically as follows: with each The polygonal framework of aggregate particles generates a rigid cluster template, which is used as the basis for generating rigid clusters with the same shape as the polygonal aggregate, and then uses the Bubble Pack algorithm to automatically fill the rigid cluster template with Pebble particles to obtain rigid clusters. 5. the generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting overlapping contact cutting algorithm according to claim 4, it is characterized in that, in described step 3, rigid cluster collision characteristic is as follows: In the discrete element software, the collision between the rigid clusters is simulated by the discrete element method. The boundary of the model is a rectangle, and the boundary of the model is set as a rigid wall, and the stiffness is given. At the same time, the rigid cluster of the aggregate is also given a contact model and Stiffness, due to the stiffness of both the wall and the aggregate, they will collide and bounce when overlapping occurs; through the mutual collision and bounce between the aggregate rigid clusters and the collision between the aggregate rigid cluster and the wall boundary, the aggregate Particle redistribution and allow for the presence of aggregate particles on the boundaries of the model. 6. the generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting overlapping contact cutting algorithm according to claim 5, is characterized in that, Set the stiffness of the model boundary wall to be much greater than that of the aggregate rigid cluster; for the aggregate particles on one side of the boundary wall, the part beyond the side boundary is placed on the other side of the boundary that is symmetrical to it through the periodic boundary; Aggregate particles located at the corner points of the model boundary are divided into four parts, which are respectively placed inside the four corresponding corner points of the model boundary through periodic boundaries. 7. The generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting the overlapping contact cutting algorithm according to claim 6, characterized in that, in the step 4, the contact overlapping judging algorithm is used to judge the overlapping of the aggregate Type, specifically as follows: read the geometry file to obtain the coordinate point information of each aggregate polygon; then call the polygon contact overlap judgment algorithm to judge the contact overlap of the aggregate polygon; The contact overlap of aggregates includes the following seven types: 1) There is point-to-point contact between adjacent polygonal aggregates; 2) There is point-to-line contact between adjacent polygonal aggregates; 3) There is line-to-line contact between adjacent polygonal aggregates; 4) There is overlap between adjacent polygonal aggregates, and the overlapping part is a polygon, and the overlapping part is inside the model boundary and has no contact with the boundary; 5) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons, and the overlapping part is inside the model boundary and has no contact with the boundary; 6) There is overlap between adjacent polygonal aggregates, and the overlapping part is a polygon, and the overlapping part is in contact with the model boundary, and there is only one intersection point between the two polygonal aggregates; 7) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons. At the same time, the overlapping part is in contact with the model boundary, and there are multiple intersection points between the two polygonal aggregates. 8. The method for generating a two-dimensional ultra-high volume fraction melange numerical simulation model using an overlapping contact cutting algorithm according to claim 7, characterized in that, in the step 5, different cutting methods are implemented according to different contact overlapping types method, as follows: 1) Point-to-point contact between adjacent polygonal aggregates, simulating the contact between sharp corners and sharp corners between aggregate particles in melange in reality, which is allowed to exist without cutting; 2) Point-to-line contact between adjacent polygonal aggregates, simulating the contact between sharp corners and surfaces between aggregate particles in melange in reality, which is allowed to exist without cutting; 3) Line-to-line contact between adjacent polygonal aggregates, simulating the surface-to-surface contact between aggregate particles in melange in reality, which is allowed and does not need to be cut; 4) There is an amount of overlap between adjacent polygonal aggregates, and the overlapping part is a polygon. The overlapping part is located inside the model boundary and has no contact with the boundary. The overlapping aggregates in this case are cut according to the following method: 4.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint For polygonal aggregates, the overlapping parts are deleted; 4.2) Adopt the geometric figure intersection discrimination method to obtain the intersection of two polygonal aggregates with overlapping amounts; read the geometric file, obtain the set of coordinates of each vertex of the polygon corresponding to the overlapping part, and record it as the first set; The set of the coordinates of each vertex of the aggregate itself is recorded as the second set; then the intersection of the first set and the second set is calculated, which is recorded as the third set; finally, the vertex coordinates in the third set are deleted from the first set, The remaining two vertices are the intersection points of the two aggregates; 4.3) Make a cutting line through these two intersection points, and cut the combined polygonal aggregate into two aggregates without overlapping along the cutting line; 5) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons. It is inside the model boundary and has no contact with the boundary. In this case, the overlapping aggregates are cut according to the following method: 5.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint For polygonal aggregates, the overlapping parts are deleted; 5.2) The intersection of two aggregates with an overlapping amount is obtained by using the geometric figure intersection discrimination method. The intersection is a plurality of polygons, and the coordinates of the intersection point cannot be obtained directly. It is necessary to further traverse the intersection to obtain the vertex coordinate set of each polygon; 5.3) Calculate the distance between any two vertex coordinate points in the intersection, find the two coordinate points with the largest distance as the intersection point of the aggregate, make a cutting line through these two intersection points, and combine polygonal aggregates along the cutting line Cut into two aggregates with no overlapping amount; 6) There is an overlapping amount between adjacent polygonal aggregates, and the overlapping part is a polygon, and at the same time the overlapping part is in contact with the model boundary, the overlapping part is divided into two parts by the boundary of this side, one part is located near the side boundary, and the other part The periodical boundary is placed near the boundary on the other side that is symmetrical to it; there is only one intersection point between the two shaped aggregates in each part, and the overlapping aggregates for this case are cut as follows: 6.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint For polygonal aggregates, the overlapping parts are deleted; 6.2) Use the geometrical figure intersection discrimination method to obtain the intersection of two aggregates with overlapping amounts. Since the overlapping part is in contact with the boundary, there is only one intersection point in the overlapping part. Use a method similar to step 4.2) to find the intersection point; then The vertex coordinates of the intersection are traversed and judged, and any vertex on the boundary is found as another intersection point; 6.3) Make a cutting line through these two intersection points, and cut the combined polygonal aggregate into two aggregates without overlapping along the cutting line; the other part of the overlapping amount located near the symmetrical boundary is cut using the same method as above; 7) There is overlap between adjacent polygonal aggregates, and the overlapping part is multiple polygons. At the same time, the overlapping part is in contact with the model boundary, then the overlapping part is divided into two parts by the boundary of this side, one part is located near the side boundary, and the other One part is placed near the border on the other side that is symmetrical to it through the periodic boundary; there are multiple intersections between the two shaped aggregates in each part, and the overlapping aggregates for this case are cut as follows: 7.1) Use the efficient joint algorithm of overlapping polygons to combine two aggregates with contact overlap, the polygons corresponding to the overlapping parts are merged, the intersecting lines are dissolved and joined at one point, and the repeated points are merged to form a joint For polygonal aggregates, the overlapping parts are deleted; 7.2) Using the intersection discrimination method of geometric figures, the intersection of two aggregates with overlapping amounts is obtained. Since the overlapping part is in contact with the boundary, there are multiple intersection points in the overlapping part, and the coordinates of the intersection point cannot be obtained directly. It is necessary to further traverse the intersection to obtain A collection of vertex coordinates for each polygon; 7.3) Calculate the distance between any two vertex coordinate points in the intersection, find the two coordinate points with the largest distance as the intersection point of the aggregate, make a cutting line through these two intersection points, and combine polygonal aggregates along the cutting line Cut into two aggregates with no overlapping amount; the other overlapping amount located near the symmetrical boundary is cut in the same way as above. 9. the generation method of the two-dimensional ultra-high volume fraction melange numerical simulation model adopting overlapping contact cutting algorithm according to claim 8, it is characterized in that, in described step 7, calculate the volume of melange model aggregate after cutting The scoring method is as follows: Step 7.1: read the geometric model file of the cut polygonal aggregate; Step 7.2: Calculate the polygonal area of each aggregate particle after cutting, and add up the total area S _a of all aggregates after cutting; Step 7.3: Calculate the ratio of the total area S _a of all aggregates after cutting to the area S _b of the boundary area of the model to obtain the volume fraction of aggregates after cutting.

Images