Simulation device, simulation method, and simulation program

Abstract

To achieve improvement in flexibility of element division that is executed in evaluation of a breakage risk of a product.SOLUTION: A simulation device includes: a modeling unit that, for each of elements into which a model of a product using a metal material is divided, models an inclusion on the basis of occurrence frequency indicating the number of inclusions that are present per unit volume and a distribution function indicating sizes of the inclusions in a probability distribution; and a simulation unit that executes a simulation of the number of times the product breaks on the basis of the inclusion model modeled for each element by the modeling unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a simulation device, a simulation method, and a simulation program. [Background technology]

[0002] One of the technologies for evaluating the risk of fatigue failure in products that use metallic materials is a component destruction evaluation device.

[0003] For example, the component destruction assessment device divides the mechanical component into a plurality of virtual cells. In this case, the mechanical component is divided into a size such that each virtual cell contains one inclusion and the volume is equally divided among the virtual cells. Then, under the assumption that the distribution function of the inclusion size follows the general Pareto distribution, the component destruction assessment device derives the probability that the stress amplitude of the acting stress exceeds the stress amplitude of the fatigue strength for each virtual cell. Based on the fatigue strength exceedance probability thus derived for each virtual cell, the component destruction assessment device derives the probability that the stress amplitude of the acting stress exceeds the stress amplitude of the fatigue strength in at least one of the virtual cells as the fatigue strength exceedance probability of the mechanical component. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-60428 [Patent Document 2] Patent No. 5445727 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the conventional technology represented by the above-mentioned component destruction evaluation device, there is room for improvement in that the flexibility of element division, which affects the risk evaluation of fatigue destruction of a product, is lost.

[0006] In one aspect, the present invention aims to provide a simulation device, a simulation method, and a simulation program that can improve the flexibility of element division performed when evaluating the risk of product breakage. [Means for solving the problem]

[0007] A simulation device according to one aspect has a modeling unit that models inclusions for each element into which a model of a product using a metallic material is divided, based on an occurrence frequency that indicates the number of inclusions present per unit volume and a distribution function that indicates the size of the inclusions as a probability distribution, and a simulation unit that executes a simulation of the number of breakages of the product based on the model of the inclusions modeled for each element by the modeling unit. Effect of the Invention

[0008] It is possible to improve the flexibility of element division performed when evaluating the risk of product breakage. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram illustrating an example of a functional configuration of a server device. [Diagram 2] FIG. 2 is a diagram showing an example of an SN diagram. [Diagram 3] FIG. 3 is a diagram showing an example of the relationship of ΔK−N / √Area. [Figure 4] FIG. 4 is a diagram illustrating an example of the analysis model. [Diagram 5] FIG. 5 is a diagram illustrating an example of the analysis model. [Figure 6] FIG. 6 is a schematic diagram showing an example of occurrence frequency. [Figure 7] FIG. 7 is a diagram illustrating an example of a distribution function. [Figure 8] FIG. 8 is a flowchart showing the procedure of the simulation process. [Figure 9] FIG. 9 is a flowchart showing the procedure of the optimization process for the inclusion distribution. [Figure 10] FIG. 10 is a diagram illustrating an example of a hardware configuration. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments of a simulation device, a simulation method, and a simulation program according to the present application will be described with reference to the accompanying drawings. Each embodiment merely shows one example or aspect, and the range of values, functions, and usage scenes are not limited by such examples. Each embodiment can be appropriately combined within a range that does not cause contradictions in the processing contents.

[0011] <Overall composition> Fig. 1 is a block diagram showing an example of a functional configuration of a server device 10. A CAE (Computer Aided Engineering) system 1 shown in Fig. 1 is an information system that provides the function of a CAE tool that supports the design and development of a product by performing analysis and simulation on a product that is virtually modeled on a computer.

[0012] As shown in Fig. 1, the CAE system 1 may include a server device 10 and a client terminal 30. Although Fig. 1 shows an example in which there is one client terminal 30, this is merely an example, and any number of client terminals 30 may be included.

[0013] The server device 10 and the client terminal 30 are communicatively connected via a network NW. The network NW may be any type of communication network, such as the Internet or a LAN (Local Area Network), regardless of whether it is wired or wireless.

[0014] The server device 10 is an example of a computer that provides the functions of the CAE tool. In one embodiment, the server device 10 can provide the CAE service by causing any computer to execute software that realizes the CAE tool service. For example, the server device 10 can be realized as a server that provides the functions of the CAE tool on-premise. In addition, the server device 10 can be realized as a PaaS (Platform as a Service) type or SaaS (Software as a Service) type application, thereby providing the functions of the CAE tool as a cloud service.

[0015] The client terminal 30 corresponds to an example of a computer that receives various functions related to the above-mentioned CAE tool. For example, the client terminal 30 is realized by a desktop or laptop personal computer. This is merely an example, and the client terminal 30 may be any computer such as a mobile terminal device or a wearable terminal.

[0016] <One aspect of the issue> As explained in the background art above, the conventional technology represented by the component destruction assessment device described above has room for improvement in terms of the flexibility of element division that affects the risk assessment of fatigue destruction of a product.

[0017] As one aspect, according to the above-mentioned component destruction evaluation device, the size of the division of the machine component into virtual cells is restricted by inclusions. That is, although countless inclusions are distributed in the material, they cannot become the origin of fatigue fracture unless they are large enough. Therefore, if the size of the virtual cells is determined based only on inclusions that can become the origin of fatigue fracture, the size of each virtual cell becomes large, and the calculation accuracy of the stress deteriorates. However, if the size of the virtual cells is set based on inclusions of a size that cannot become the origin of fatigue fracture in order to increase the calculation accuracy of the stress, the number of meshes and mesh density increase, and the calculation cost also increases.

[0018] As a further aspect, the above-mentioned component destruction evaluation device imposes a constraint that the volume of each virtual cell must be divided equally. Due to such constraint, it is difficult to realize element division that changes the mesh density of the virtual cells. For example, it is difficult to increase the mesh density of a portion that may be the starting point of fatigue destruction, such as a portion where stress concentration is likely to occur, compared to other portions. For this reason, it becomes more difficult to apply the method to a product as the shape of the product becomes more complex. Furthermore, since the entire product is evenly divided into high-density virtual cells during element division, the overall calculation cost increases. In addition, the calculation cost of a portion that may not be the starting point of fatigue destruction increases, so the unnecessary calculation cost also increases.

[0019] <One aspect of the problem-solving approach> Therefore, the CAE tool according to this embodiment is equipped with a virtual fatigue testing function for simulating fatigue testing of the product. That is, the virtual fatigue testing function simulates fatigue testing of the product by modeling an inclusion model for each element into which the product is divided during analysis, based on an occurrence frequency indicating the number of inclusions present per unit volume and a distribution function indicating the size of the inclusions as a probability distribution.

[0020] Such a virtual fatigue testing function can quantitatively evaluate the risk of breakage of a product, such as the number of breakages and the starting point of breakage. In this case, the virtual fatigue testing function can suppress restrictions on the size of the division of the product into elements due to inclusions and restrictions such as the requirement to divide the volume of each element equally. Therefore, the virtual fatigue testing function according to this embodiment can improve the flexibility of the element division performed when evaluating the risk of breakage of a product.

[0021] <Server device configuration> Fig. 1 shows a schematic diagram of blocks related to the functions of the CAE tool included in the server device 10. As shown in Fig. 1, the server device 10 includes a communication control unit 11, a storage unit 13, and a control unit 15. Note that Fig. 1 shows only a selection of functional units related to the functions of the CAE tool, and the server device 10 may include functional units other than those shown in the figure.

[0022] The communication control unit 11 is a functional unit that controls communication with other devices such as the client terminal 30. For example, the communication control unit 11 is realized by a network interface card such as a LAN card. As merely one example, the communication control unit 11 can receive requests such as a request to execute a virtual fatigue test from the client terminal 30, or output the execution results of the virtual fatigue test.

[0023] The storage unit 13 is a functional unit that stores various data. As just one example, the storage unit 13 is realized by an internal, external, or auxiliary storage of the server device 10. For example, the storage unit 13 stores inclusion setting data 13A. Data other than the inclusion setting data 13A, such as a 3D (dimensions) model of a product, analysis condition settings, mesh division settings, and the like, may be stored in the storage unit 13.

[0024] The control unit 15 is a processing unit that performs overall control of the server device 10. For example, the control unit 15 can be realized by a hardware processor. As shown in Fig. 1, the control unit 15 has a receiving unit 15A, an analyzing unit 15B, a modeling unit 15C, and a simulation unit 15D. The control unit 15 may be realized by hardwired logic.

[0025] The reception unit 15A is a processing unit that receives various requests from the client terminal 30. For example, the reception unit 15A can receive a request to execute a virtual fatigue test. Before or after or at the same time as the request to execute such a virtual fatigue test, the reception unit 15A can receive designation of a product to be subjected to the virtual fatigue test and furthermore, settings of analysis conditions for the product.

[0026] For example, the reception unit 15A can receive, as a product specification, a specification of a product model number or a selection from a list of 3D models of the product. Hereinafter, a product model, for example, a 3D model or a 2D model, may be referred to as a "product model."

[0027] Furthermore, the accepting unit 15A can accept, as examples of analysis conditions, material property values ​​of a product, such as Young's modulus and Poisson's ratio, as well as material strength, boundary conditions, and the like.

[0028] For example, material strength can be modeled as an S (Stress)-N (Number of cycles to failure) diagram. FIG. 2 is a diagram showing an example of an SN diagram. FIG. 2 shows an SN diagram G1 of a steel material used as a material of a compression coil spring as an example of a product to be input into a CAE tool. As shown in FIG. 2, material strength can be modeled by the SN diagram G1 showing the relationship between the load (stress) S applied to the material and the number of loads until the material breaks, so-called number of cycles to failure N. The SN diagram G1 can be modeled as a curve in which the number of cycles to failure N increases as the stress S decreases, and the number of cycles to failure N decreases as the stress S increases. Furthermore, the SN diagram G1 also models a lower limit value of stress S, so-called fatigue limit, at which fatigue failure does not occur no matter how many times the load is repeated, by bending a straight line or curve sloping downward to the right and becoming horizontal or approximately horizontal.

[0029] In addition to this SN diagram, material strength can also be modeled based on the relationship between the size of the inclusion, the stress intensity factor calculated from the stress, and the number of repeated fractures. The "stress intensity factor" referred to here can be calculated according to the following formula (1), which is merely one example. The "σ ai " refers to the stress amplitude at the inclusion position where the crack started, taking into account the effect of the load stress gradient. For example, in an axial load fatigue test, σ a The rotating bending fatigue test is σ a (1-2d inc / d). Note that σ a refers to the nominal stress amplitude on the surface of the specimen, d refers to the minimum cross-sectional diameter of the specimen, and d inc indicates the distance from the test piece surface to the center of the inclusion.

[0030]

number

[0031] FIG. 3 is a diagram showing an example of the relationship between ΔK-N / √Area. The vertical axis of graph G2 shown in FIG. 3 indicates N / √Area, and the horizontal axis indicates ΔK. The inclusion size √Area given here indicates the square root of the area when the inclusion is projected in the principal stress direction. As shown in FIG. 3, graph G2 can be modeled as a straight line sloping downward to the right. According to such graph G2, it is clear that the number of cycles to fracture N decreases as the stress intensity factor ΔK increases, while the number of cycles to fracture N increases as the stress intensity factor ΔK decreases.

[0032] When a product breaks due to fatigue caused by an inclusion in the material, the fatigue strength changes depending on the size of the inclusion in the material, but the size of the inclusion is not uniform. For this reason, it is difficult to evaluate the randomness of inclusion size using an SN diagram that uses stress and the number of repeated fractures as evaluation axes. On the other hand, ΔK-N / √Area makes it possible to evaluate the randomness of inclusion size, so that the fatigue strength of the material when fatigue fracture originates from an inclusion can be modeled more precisely.

[0033] 2 and 3 show graphs of material strength, but modeling may be performed using a function showing the relationship between stress S and the number of fracture cycles N, or a function showing the relationship between inclusion size √Area, stress intensity factor ΔK, and the number of fracture cycles N. Also, while Fig. 3 shows an example of modeling the relationship ΔK-N / √Area including inclusions, by formulating it including various breakage factors other than inclusions (surface breakage, crystal cracks, etc.), it becomes possible to predict not only the number of breakages but also the origin and cause of breakage.

[0034] As an example of other analysis conditions, the reception unit 15A can receive settings of boundary conditions such as load conditions and constraint conditions. For example, constraint conditions similar to the environment in which the product is actually used are set, and load conditions similar to the load actually applied in the environment are set. As an example only, when a virtual fatigue test of a compression coil spring is performed, a constraint condition for fixing one of both ends of the compression coil spring can be set, and a load corresponding to a stroke in the compression direction of the compression coil spring from the other end can be set as a load condition.

[0035] The analysis unit 15B is a processing unit that executes an analysis of a product model. As merely one example, the analysis unit 15B executes a numerical analysis of the product model according to an algorithm such as finite element analysis, so-called FEM (Finite Element Method), FEA (Finite Element Analysis), the boundary element method, or the finite volume method.

[0036] More specifically, the analysis unit 15B acquires a product model corresponding to the product specification accepted by the acceptance unit 15A. As an example of such a product model, the analysis unit 15B can acquire CAD (Computer-Aided Design) data in which the shape of the product is modeled.

[0037] Next, the analysis unit 15B divides the product model into a plurality of elements. The "elements" referred to here may also be called "cells," "virtual cells," "meshes," or "mesh elements." For example, the analysis unit 15B refers to the shape of the elements of the analysis model obtained by the mesh division and the settings of the size and density of the mesh dividing the product model. Then, the analysis unit 15B divides the product model into meshes according to the settings of the shape of the elements and the size and density of the mesh. As a result of such mesh division, the product model is divided into tetrahedral elements, hexahedral elements, etc., and a set of tetrahedral elements and hexahedral elements is obtained as the analysis model.

[0038] Figures 4 and 5 are diagrams showing an example of an analysis model. Figures 4 and 5 show analysis models A1 and A2 in which a compression coil spring is divided into meshes as an example of a product model. Furthermore, Figure 5 shows cross sections A21 and A22 as examples of cross sections of the third winding from the start of winding of a compression coil spring.

[0039] As shown in FIG. 4, according to the analysis model A1, the volume size is divided equally among the elements, and the mesh density is uniformly divided throughout the product model of the compression coil spring. On the other hand, as shown in FIG. 5, according to the analysis model A2, the mesh size is finer than that of the analysis model A1 shown in FIG. 4, or the mesh density is higher than that of the analysis model A1. Furthermore, as shown in the cross section A21, the volume and density can be divided equally among the elements, and as shown in the cross section A22, the volume and density can be changed between the elements closer to the surface and the elements closer to the center. As one aspect, in the case of a spring, stress tends to occur more easily as it approaches the surface, and less easily as it approaches the center. Therefore, by implementing the element division shown in the cross section A22, calculation resources can be allocated to the locations where stress concentration may occur, making it possible to make the calculation more efficient.

[0040] In this way, the analysis unit 15B can set the mesh size finer or the mesh density higher for a part of the product model where the shape change is large or where stress concentration may occur. Such mesh size and mesh density settings may be realized manually by user definition, or may be realized automatically as a function of a CAE tool. In the latter case, as an example only, a template in which shape change where stress concentration is likely to occur is standardized is matched with a part cut out from a product model while changing its size and range, thereby extracting a part whose similarity is equal to or higher than a threshold value. The mesh size of the part extracted in this way can be set finer than other parts, or the mesh density can be set higher than other parts.

[0041] After the element division, the analysis unit 15B executes a numerical analysis to calculate the volume and stress of each element into which the product model is divided, based on the analysis conditions received by the reception unit 15 A. By such a numerical analysis, statistics or distribution of the volume and stress displaced by one cycle of shape change are calculated.

[0042] The modeling unit 15C is a processing unit that models inclusions based on the inclusion setting data 13A. Here, the inclusion setting data 13A includes an "occurrence frequency" indicating the number of inclusions present per unit volume and a "distribution function" indicating the probability distribution of inclusion sizes.

[0043] For example, the occurrence frequency is expressed as the unit volume V [mm 3 ] is defined as the number A of inclusions present per unit volume V [mm ]. FIG. 6 is a schematic diagram showing an example of the occurrence frequency. FIG. 6 shows a schematic representation of the occurrence frequency when A / V=6. As shown in FIG. 6, when A / V=6, 3 ] contains six inclusion models, M1 to M6.

[0044] Moreover, the distribution function is defined by the inclusion size and the occurrence probability of the inclusions. FIG. 7 is a diagram showing an example of the distribution function. The horizontal axis of the graph shown in FIG. 7 indicates the inclusion size, and the vertical axis indicates the occurrence probability of the inclusions. In the example shown in FIG. 7, a graph G3 of a distribution function in which the probability distribution of the inclusion size is approximated to a general Pareto distribution is shown as merely an example. Note that this is merely an example, and the distribution function is not limited to the general Pareto distribution. For example, the distribution function may be approximated to other probability density functions such as a power distribution or a power law distribution in addition to the general Pareto distribution.

[0045] As an example, the modeling unit 15C executes the following process for each element into which the product model is divided. That is, the modeling unit 15C calculates the number of inclusions present in the element based on the occurrence frequency included in the inclusion setting data 13A. For example, if the volume of the element obtained as a result of the numerical analysis by the analysis unit 15B is VE [mm 3 ], the probability that an inclusion exists in an element can be calculated by VE*A / V. The number of inclusions in an element is calculated by adding the number corresponding to the integer part of the VE*A / V numerical sequence calculated in this way to the result of a lottery that determines whether or not an inclusion exists with a probability corresponding to the decimal part.

[0046] For example, take the case where VE*A / V=0.67. In this case, since the numerical value of the integer part is zero, the modeling unit 15C executes only the lottery corresponding to the numerical value of the decimal part. That is, the modeling unit 15C generates random numbers with a probability of 67% corresponding to the numerical value of the decimal part "0.67" in the numerical range of the random number generation, that is, about two-thirds of the numerical value is a winning number with inclusions, and about one-third of the numerical value is a losing number without inclusions. At this time, if the random number is a winning number, the modeling unit 15C calculates the number of inclusions in the element as "1" by adding the number "0" corresponding to the numerical value of the integer part and the lottery result "1" corresponding to the numerical value of the decimal part. On the other hand, if the random number is not a winning number, the modeling unit 15C calculates the number of inclusions in the element as "0" by adding the number "0" corresponding to the numerical value of the integer part and the lottery result "0" corresponding to the numerical value of the decimal part.

[0047] Also, take the case where VE*A / V=1.67 as an example. In this case, since the numerical value of the integer part is "1", the modeling unit 15C identifies the number corresponding to the numerical value of the integer part as "1". Furthermore, the modeling unit 15C generates random numbers with a probability of 67% corresponding to the numerical value of the decimal part "0.67", that is, approximately two-thirds of the numerical value, as a winning number with inclusions, and approximately one-third of the numerical value as a losing number without inclusions. At this time, if the random number is a winning number, the modeling unit 15C calculates the number of inclusions in the element as "2" by adding the number "1" corresponding to the numerical value of the integer part and the lottery result "1" corresponding to the numerical value of the decimal part. On the other hand, if the random number is not a winning number, the modeling unit 15C calculates the number of inclusions in the element as "1" by adding the number "1" corresponding to the numerical value of the integer part and the lottery result "0" corresponding to the numerical value of the decimal part.

[0048] After calculating the number of inclusions in the element, the modeling unit 15C determines the inclusion size for each inclusion in the element based on the distribution function included in the inclusion setting data 13A. For example, the modeling unit 15C assigns an inclusion size label to each numerical value included in the numerical range of the random number generation according to a ratio corresponding to the occurrence probability of the inclusion. Then, the modeling unit 15C generates a random number and determines the label assigned to the generated random number as the inclusion size.

[0049] The simulation unit 15D is a processing unit that simulates a fatigue test of a product. As an example, the simulation unit 15D judges whether or not an inclusion exists in an element, that is, whether or not the number of inclusions in the element is 1 or more. At this time, if the number of inclusions in the element is 1 or more, the simulation unit 15D selects an inclusion that has the largest inclusion size determined by the modeling unit 15C among the inclusions included in the element. Based on the inclusion size of the inclusion selected in this way and the stress intensity factor obtained from the stress of the element among the stresses calculated for each element by the analysis unit 15B, the simulation unit 15D calculates the number of times the element breaks due to fatigue. Hereinafter, the number of times the element breaks due to fatigue may be referred to as the "number of breaks". In this case, the number of breaks can be derived from the relational expression ΔK-N / √Area shown in FIG. 3. On the other hand, if the number of inclusions in the element is not 1 or more, the simulation unit 15D calculates the number of breaks of the element based on the statistical value of the stress calculated by the analysis unit 15B, for example, the stress amplitude. In this case, the number of breakages can be derived from the SN diagram shown in FIG.

[0050] After calculating the number of breakages of each element, the simulation unit 15D identifies the minimum number of breakages among all the elements as the number of breakages of the product, and identifies the element among all the elements for which the minimum number of breakages is observed as the breakage starting position of the product.

[0051] In this way, the virtual fatigue test simulating the number of breakages and the breakage initiation positions of the product is repeated a specific number of tests K. For example, the number of tests K may be set to the number of product shipments or the number of product shipments plus a safety margin from the perspective of product quality assurance. After that, the simulation unit 15D calculates a statistical value of the number of product breakages and a statistical value of the product breakage initiation positions during the K tests. Examples of such statistical values ​​include at least one of the average value, median value, mode value, variance, standard deviation, etc.

[0052] The statistical values ​​of the number of breakages and the breakage initiation positions of these products can be output as simulation results to any output destination. For example, the simulation results can be output to the client terminal 30. In addition, the simulation results can be output to software or services that realize various processes based on the simulation results, such as a process for determining the price of a product or a process for analyzing the fatigue strength of an assembled product generated by assembling the product including the product.

[0053] <Processing flow> 8 is a flowchart showing the procedure of a simulation process, which is merely an example and can be executed when a request to execute a virtual fatigue test is received by the receiving unit 15A.

[0054] 8, the analysis unit 15B acquires a product model corresponding to the product specification accepted by the acceptance unit 15A (step S101). Next, the analysis unit 15B divides the product model acquired in step S101 into a mesh, thereby dividing the product model into a plurality of elements (step S102).

[0055] Then, the analysis unit 15B executes a numerical analysis to calculate the volume and stress of each element divided in step S102 based on the analysis conditions received by the reception unit 15A (step S103).

[0056] Thereafter, loop process 1 is executed in which the processes from step S104 to step S109 are repeated a number of times corresponding to a specific number of tests K. Note that, although Fig. 8 shows an example in which the processes from step S104 to step S109 are executed as loop process 1, the processes from step S104 to step S109 do not necessarily have to be executed serially, and may be executed in parallel for every K tests.

[0057] Furthermore, in the K virtual fatigue tests executed in the above loop process 1, a loop process 2 is executed in which the processes from step S104 to step S108 are repeated a number of times corresponding to the number L of elements included in the product model. Note that, although an example in which the processes from step S104 to step S108 are executed as loop process 2 is given in Fig. 8, the processes from step S104 to step S108 do not necessarily have to be executed serially, and may be executed in parallel for each L elements.

[0058] That is, the modeling unit 15C calculates the number of inclusions present in the element based on the occurrence frequency included in the inclusion setting data 13A (step S104).

[0059] At this time, if an inclusion is present in the element, i.e., if the number of inclusions in the element is 1 or more (Yes in step S105), loop process 3 is executed in which the process of step S106 is repeated a number of times corresponding to the number M of inclusions. Note that, although an example is given in Fig. 8 in which the process of step S106 is executed as loop process 3, the process of step S106 does not necessarily have to be executed serially, and may be executed in parallel for each M inclusions.

[0060] That is, the modeling unit 15C determines the inclusion size of the inclusions based on the distribution function included in the inclusion setting data 13A (step S106).

[0061] By repeating such loop process 3, it is possible to determine the inclusion size for each of the M inclusions calculated as the number of inclusions in the element in step S104.

[0062] Next, simulation unit 15D selects the inclusion having the largest inclusion size determined in step S106 from among the M inclusions (step S107). Based on the inclusion size of the inclusion selected in step S107 and the stress intensity factor obtained from the stress of the element among the stresses calculated for each element in step S103, simulation unit 15D calculates the number of breakages of the element (step S108).

[0063] By repeating such loop process 2, the number of breakages can be calculated for each of the L elements divided in step S102.

[0064] Thereafter, the simulation unit 15D identifies the minimum number of breakages among the number of breakages calculated for each of the L elements as the number of breakages of the product, and identifies the element among the L elements in which the minimum number of breakages is observed as the breakage starting position of the product (step S109).

[0065] By repeating this loop process 1, the number of breakages and the breakage starting points of the product can be identified for each K tests.

[0066] Then, the simulation unit 15D calculates the statistical value of the number of breakages of the product and the statistical value of the breakage starting point positions during the K tests (step S110), and ends the process.

[0067] <One aspect of the effect> As described above, the server device 10 according to the present embodiment simulates a fatigue test of a product by modeling a model of an inclusion for each element into which the product is divided during analysis based on the occurrence frequency indicating the number of inclusions present per unit volume and the distribution function indicating the size of the inclusions as a probability distribution. As a result, as one aspect, the breakage risk of the product, for example, the number of breakages and the breakage starting point, can be quantitatively evaluated. Furthermore, the range in which the breakage risk can be evaluated is not limited to high cycles, but can be expanded to low cycles and ultra-high cycles. Therefore, the optimization of the product shape and the selection of the material to be used can be performed more effectively. As another aspect, the size of the product divided into elements can be suppressed from being restricted by the inclusions, or from being restricted such that the volume of each element must be divided evenly. Therefore, according to the server device 10 according to the present embodiment, it is possible to improve the flexibility of the element division performed when evaluating the breakage risk of the product.

[0068] <Application Examples> The above-described embodiment is merely an example, and various applications are possible.

[0069] For example, the occurrence frequency and distribution function included in the inclusion setting data 13A can be optimized by a simulation process shown in FIG.

[0070] FIG. 9 is a flowchart showing the procedure of the optimization process of the inclusion distribution. As shown in FIG. 9, the control unit 15 acquires the results of the fatigue test of the material (step S301). The "fatigue test of the material" referred to here is a test that is not virtual but is actually performed, and may be distinguished from the above-mentioned virtual fatigue test. A rotating bending fatigue test having a large risk volume is merely one example of such a fatigue test of the material. Furthermore, the results of the fatigue test of the material may include the number of breaks, the break initiation position, for example, the depth from the surface, and the inclusion size at the break initiation position.

[0071] Next, the control unit 15 sets the parameters that define the distribution function and the initial values ​​of the occurrence frequency (step S302). The values ​​set in this step S302 are merely "initial values" and do not necessarily have to be exact numerical values ​​that correspond to the actual distribution of inclusions. For example, when approximating the distribution function to a general Pareto distribution, the initial values ​​of the parameters of the distribution function can be determined by randomly setting the position parameter, scale parameter, and shape parameter that define the general Pareto distribution. The initial value of the occurrence frequency can also be determined by randomly setting the number A of the occurrence frequency A / V.

[0072] Then, the control unit 15 executes a loop process 1 in which the processes from step S303 to step S305 described below are repeated until a convergence condition is satisfied under which the optimization of the parameters of the distribution function and the occurrence frequency converges.

[0073] That is, the control unit 15 executes a simulation process shown in Fig. 8 for a product corresponding to the fatigue test of the material acquired in step S301 (step S303). In the simulation process in step S303, the initial value of the parameter of the distribution function set in step S302 when calculating the number of inclusions in step S104 shown in Fig. 8, or the parameter of the distribution function updated in step S305 described later, is used. Furthermore, in the simulation process in step S303, the initial value of the occurrence frequency set in step S302 when calculating the inclusion size in step S106 shown in Fig. 8, or the occurrence frequency updated in step S305 described later, is used.

[0074] Then, the control unit 15 compares the result of the fatigue test of the material acquired in step S301 with the result of the simulation executed in step S303 (step S304). As an example only, the control unit 15 calculates the difference in the number of breakages, the difference in the size of the inclusions at the breakage initiation points, and the difference in the breakage initiation points positions between the result of the fatigue test of the material and the result of the simulation.

[0075] Thereafter, the control unit 15 executes parameter update for optimizing an objective function including the difference in the number of breakages, the difference in the size of the inclusion at the breakage starting point, and the difference in the breakage starting point position, in this example, updating the parameters of the distribution function and the occurrence frequency (step S305). As merely an example of the difference, the parameters of the distribution function and the occurrence frequency can be optimized by parameter update for minimizing a loss function including the loss in the number of breakages, the loss in the size of the inclusion at the breakage starting point, and the loss in the breakage starting point position. It is also possible to use similarity or correlation coefficient as a likelihood instead of this loss, and optimize the parameters of the distribution function and the occurrence frequency by maximizing the likelihood or minimizing the negative log likelihood.

[0076] Such loop process 1 can be executed a predetermined number of times or until the change in the parameter updated in step S305 becomes less than a threshold value. When the convergence conditions, such as the condition for the number of executions and the condition for the parameter change amount, are satisfied, loop process 1 can be terminated.

[0077] Then, the control unit 15 stores the parameters of the distribution function and the occurrence frequency obtained by satisfying the above-mentioned convergence condition in the inclusion setting data 13A of the storage unit 13 (step S306), and ends the process.

[0078] In this way, the occurrence frequency and distribution function of inclusions are optimized based on the results of comparing the number of breaks, the break initiation positions, and the size of inclusions at the break initiation positions obtained as a result of the fatigue test of the material with the number of breaks, the break initiation positions, and the size of inclusions at the break initiation positions obtained as a result of the virtual fatigue test. Through such optimization, the occurrence frequency and distribution function of inclusions can be inversely estimated.

[0079] The cleanliness of metal materials can be evaluated based on the occurrence frequency and distribution function of these inclusions, providing evidence that is useful for selecting materials to be used in products. In addition, product fatigue testing can be performed using virtual fatigue testing, reducing implementation costs.

[0080] Furthermore, compared to a method of determining the maximum inclusion within a usable range by extreme value statistics based on inclusion information obtained by a SEM (Scanning Electron Microscope) or fatigue testing of materials, and further compared to a conventional technique of obtaining inclusion information by melting or the like to determine the distribution, the present embodiment has the following advantageous effects: For example, the inverse estimation of inclusion distribution according to this embodiment is not restricted by the theoretical maximum inclusion, and allows a fatigue test of a product to be simulated using the occurrence frequency and distribution function of inclusions that can best reproduce the results of fatigue testing of materials, thereby preventing excessive safety design.

[0081] In addition, compared to the conventional technology of obtaining inclusion information by melting or the like to determine the distribution, the present invention has the following advantageous effects. For materials with high cleanliness, even if melting or the like is performed as in the conventional technology, inclusions of a size that would cause problems during mass production are rarely obtained. According to the inverse estimation of inclusion distribution according to the present embodiment, information on inclusions, for example, the occurrence frequency and distribution function of inclusions, can be obtained without performing such a great deal of work, for example, melting, which requires a lot of time and effort.

[0082] <Modification> Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

[0083] In addition, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads, usage conditions, etc.

[0084] Furthermore, the effects of each embodiment described in this specification are merely examples and are not limiting, and other effects may also be provided.

[0085] <Hardware configuration> Moreover, the various processes described in the above embodiment can be realized by executing a prepared program on a computer such as a personal computer or a workstation. Therefore, an example of a computer that executes a simulation program having the same functions as those in the above embodiment will be described below with reference to FIG.

[0086] Fig. 10 is a diagram showing an example of a hardware configuration. As shown in Fig. 10, the computer 100 has an operation unit 110a, a speaker 110b, a camera 110c, a display 120, and a communication unit 130. Furthermore, the computer 100 has a CPU 150, a ROM 160, a HDD 170, and a RAM 180. These units 110 to 180 are connected via a bus 140.

[0087] As shown in Fig. 10, the HDD 170 stores a simulation program 170a that performs the same functions as the reception unit 15A, the analysis unit 15B, the modeling unit 15C, and the simulation unit 15D shown in the above embodiment. This simulation program 170a may be integrated or separated like the components of the reception unit 15A, the analysis unit 15B, the modeling unit 15C, and the simulation unit 15D shown in Fig. 1. That is, the HDD 170 does not necessarily have to store all the data shown in the above embodiment 1, and it is sufficient that the data used for processing is stored in the HDD 170.

[0088] Under such an environment, the CPU 150 reads out the simulation program 170a from the HDD 170 and loads it in the RAM 180. As a result, the simulation program 170a functions as a simulation process 180a as shown in FIG. 10. The simulation process 180a loads various data read out from the HDD 170 in an area of ​​the storage area of ​​the RAM 180 that is assigned to the simulation process 180a, and executes various processes using the loaded various data. For example, the processes shown in FIG. 8 and FIG. 9 may be included as examples of the processes executed by the simulation process 180a. Note that in the CPU 150, all of the processing units shown in the above-mentioned first embodiment do not necessarily need to operate, and it is sufficient that the processing units corresponding to the processes to be executed are virtually realized.

[0089] The simulation program 170a does not necessarily have to be stored in the HDD 170 or the ROM 160 from the beginning. For example, the simulation program 170a is stored in a "portable physical medium" such as a flexible disk, a so-called FD, a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card that is inserted into the computer 100. The computer 100 may then acquire and execute the simulation program 170a from the portable physical medium. The simulation program 170a may also be stored in another computer or a server device connected to the computer 100 via a public line, the Internet, a LAN, a WAN, or the like. The simulation program 170a thus stored may be downloaded to the computer 100 and then executed. [Explanation of symbols]

[0090] 1. CAE System 10. Server equipment 11 Communication control section 13 Storage section 13A Inclusion setting data 15 Control section 15A Reception 15B Analysis Department 15C Modeling Department 15D Simulation Department 30 Client Terminals

Claims

1. a modeling unit that models the inclusions for each element into which a model of a product using a metallic material is divided, based on an occurrence frequency that indicates the number of inclusions present per unit volume and a distribution function that indicates the size of the inclusions as a probability distribution; a simulation unit that executes a simulation of the number of breakages of the product based on a model of the inclusion modeled for each of the elements by the modeling unit; A simulation device comprising:

2. the modeling unit calculates the number of inclusions in the element based on the occurrence frequency, and determines a size of each of the calculated number of inclusions based on the distribution function; the simulation unit selects an inclusion with a maximum size from among the inclusions included in the element, and calculates the number of breakages of the element based on the size of the selected inclusion and the stress generated in the element; 2. The simulation device according to claim 1.

3. The simulation unit calculates the number of breakages of the element based on a relational expression in which a relationship between stress, a size of an inclusion, and a number of repeated breakages is modeled.

3. The simulation device according to claim 2.

4. the simulation unit identifies the minimum number of breakages among the numbers of breakages calculated for each element included in the model of the product as the number of breakages of the product; 4. The simulation device according to claim 2 or 3.

5. The simulation unit identifies an element having the minimum number of breakages among elements included in the model of the product as a breakage initiation position of the product.

5. The simulation device according to claim 4.

6. The method further includes an optimization unit that optimizes an occurrence frequency and a distribution function of the inclusions based on a comparison result between the number of breaks, the break initiation position, and the size of the inclusion at the break initiation position obtained as a result of a fatigue test of the metallic material, and the number of breaks, the break initiation position, and the size of the inclusion at the break initiation position obtained as a result of a simulation by the simulation unit.

6. The simulation device according to claim 1, wherein the first and second inputs are input to the first and second inputs.

7. the optimization unit repeatedly updates the distribution function and occurrence frequency of the inclusions so as to minimize a loss function including a loss in the number of breakages, a loss in the breakage initiation position, and a loss in the size of the inclusion at the breakage initiation position, which are obtained between the results of the fatigue test of the metallic material and the results of the simulation by the simulation unit; 7. The simulation device according to claim 6.

8. For each element into which a model of a product using a metallic material is divided, a model of the inclusions is formed based on an occurrence frequency indicating the number of inclusions present per unit volume and a distribution function indicating the size of the inclusions as a probability distribution; A simulation of the number of breakages of the product is performed based on the model of the inclusions modeled for each element. A simulation method, characterized in that the processing is executed by a computer.

9. For each element into which a model of a product using a metallic material is divided, a model of the inclusions is formed based on an occurrence frequency indicating the number of inclusions present per unit volume and a distribution function indicating the size of the inclusions as a probability distribution; A simulation of the number of breakages of the product is performed based on the model of the inclusions modeled for each element. A simulation program that causes a computer to execute a process.

Images