TWI715342B - Simulation device, simulation program and simulation method - Google Patents

Abstract

[Question] The object of the present invention is to provide a technology that can appropriately support the improvement of performance due to design. [Solution] In order to solve the above-mentioned problems, one of the representative simulation devices of the present invention is provided with: a reference model memory unit as a processing object, and the reference model is memorized; and a reference model memory unit as a reference The comparison object of the model, and the reference model with performance indicators as the plural memory; and the analysis processing unit. This analysis processing unit is provided with: a space selection unit, which sets the corresponding area of each reference model for the division area of the reference model; and an analysis index extraction unit, which is for each reference model, And the difference between the distinguishing area of the reference model and the corresponding area of the reference model is obtained as the analysis index; and the analysis part is for the distinguishing area of the reference model, and for each reference model. Find the extracted analysis index and the performance index of each reference model to find the correlation.

Description

Simulation device, simulation program and simulation method

The present invention relates to a simulation device, a simulation program and a simulation method.

In the transportation system of automobiles, etc., measures have been implemented to promote energy saving by realizing the surface shape of the structure that reduces the running resistance. In addition, in automobiles, etc., measures have been implemented to improve safety and energy saving by achieving high rigidity and lightweight component shapes at the same time. These measures generally utilize CAE (Computer Aided Engeneering) analysis to improve performance.

For example, in Patent Document 1, it is disclosed that "parts of a structure composed of planar elements and/or three-dimensional elements are divided into plural parts in a uniaxial direction. For each divided part , The shape of the part is set by changing the height of the cross-section, etc.. In the state where the set parts are assembled in the aforementioned structure, the rigidity analysis of the plural types of the aforementioned structure is performed. Based on the result of the rigidity analysis Perform a multivariate analysis. Based on the results of this multivariate analysis, a rigid analysis is selected. A technique to determine the content of the section shape based on the multiple regression coefficients of the selected rigid analysis. [Prior Technical Literature] [Patent Literature]

[Patent Document 1] JP 2013-218652 A

[The problem to be solved by the invention]

Although the technology of Patent Document 1 uniformly analyzes planar elements and/or three-dimensional elements, it does not consider the narrowing of the range to places with important shapes that affect performance. In addition, the technology of Patent Document 1 has a problem that it is difficult to implement complex shapes that cannot select design parameters that affect performance.

Furthermore, in Patent Document 1, the calculation load in the performance analysis of CAE is large (for example, when performing large-scale fluid analysis, etc.), and only once for each divided part When the CAE performance analysis is performed densely and repeatedly with a large number of combinations with small changes, there is a problem that it will take an extremely long time until the best solution is obtained.

Therefore, the purpose of the present invention is to provide a technology that can efficiently support the performance improvement caused by the design. [Means to solve the problem]

In order to solve the above-mentioned problems, one of the representative simulation devices of the present invention is provided with: a reference model memory unit as the processing object, and the reference model is memorized; and a reference model memory unit as the comparison object of the reference model , And use the reference model with performance indicators as the memory of the plural; and the analytical processing unit. This analysis processing unit is provided with: a space selection unit, which sets the corresponding area of each reference model for the division area of the reference model; and an analysis index extraction unit, which is for each reference model, And the difference between the distinguishing area of the reference model and the corresponding area of the reference model is obtained as the analysis index; and the analysis part is for the distinguishing area of the reference model, and for each reference model. Find the extracted analysis index and the performance index of each reference model to find the correlation. [Effects of Invention]

In the present invention, it becomes possible to efficiently support the performance improvement caused by the design. Problems, structures, and effects other than those described above will become clearer based on the description of the following embodiments.

Hereinafter, the embodiments of the present invention will be described with reference to the drawings. In addition, in each figure, the same reference numerals are attached to the constituent elements having common functions, and the repeated description of these elements is omitted. [Example 1]

In the first embodiment, an explanation is given on the "simulation technology that uses the analysis results related to the performance improvement of the design to provide support for the design work of the user who is the designer".

<Configuration of Example 1> FIG. 1 is a diagram showing the structure of the simulation device 1 of the first embodiment. In the figure, the simulation device 1 is configured as an "information processing device (computer, etc.) provided with a CPU (Central Processing Unit) and a memory as hardware". By using this hardware to implement a simulation program, various functions described later can be realized. For part or all of this hardware, it can also be replaced by DSP (Digital Signal Processor), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit), etc. In addition, part or all of the hardware can also be centralized at a server on the network, or distributed and configured in the cloud, so that multiple people can use it together via the network.

The simulation device 1 includes a model management unit 10, an analysis processing unit 20, and an operation unit 40.

The model management unit 10 includes a reference model storage unit 11 and a reference model storage unit 12. This reference model storage unit 11 stores a data group (hereinafter referred to as a "reference model") that defines at least the design data of the shape for the object whose shape the user wants to design. The reference model memory unit 12 is used as a comparison object of the reference model, and the reference model with the performance index is used as a plural memory.

The analysis processing unit 20 includes a model setting unit 21, a space selection unit 22, an analysis index extraction unit 23, an analysis unit 24, and an analysis result display unit 26.

The model setting unit 21 acquires a set of reference models to be processed from the reference model storage unit 11. Furthermore, the model setting unit 21 acquires a plurality of reference models from the reference model storage unit 12 that can be compared with the acquired reference model.

The space selection unit 22 divides the reference model of the processing target into plural parts and sets the division area. Furthermore, the space selection unit 22 serves as an area corresponding to the division area of the reference model, and sets a corresponding area for each of the reference models.

The analysis index extraction part 23 is to obtain the difference of design data between the distinguished area of the reference model and the corresponding area of the reference model, and use it as an analysis index.

The analysis part 24 is for each of the divisional areas of the reference model to obtain correlations for "how much influence the analysis index will have on the performance index of the reference model" and "what tendency or relationship exists" .

The analysis result display unit 26 is based on the correlation of each of the divided areas, and is performed relative to the graphic display of the reference model for the "how much influence the divided area of the reference model will have on the performance" Visualize and display the color distinction. In particular, the distinguished areas that are highly correlated with performance are displayed to the user by color, area division, marking, or flashing, etc., in a way that will become conspicuous.

The operation unit 40 includes a reference model registration unit 41, a reference model registration unit 42, and a parameter registration unit 43, and provides a user interface for the user to register each of the reference model, reference model, and analysis parameters. <Action of Example 1> FIG. 2 is a flowchart illustrating the operation of the simulation device 1. Follow the step numbers in the figure to describe the operation of the simulation device 1.

Step S101: The user performs the operation of designing the shape of the object through the reference model registration unit 41, and creates a data file of the reference model in the reference model storage unit 11. In the data file of this reference model, the design data related to the shape is, for example, a collection of point coordinate data stored in a 3-dimensional space. Furthermore, in this reference model, in addition to the design data of the shape, the design related to various physical characteristics such as material, surface roughness, strength, elasticity, density, weight, etc. can also be defined for the object. data. In addition, this reference model is basically the same data format as the reference model, and it can also add a data area for storing the performance indicators described later.

Step S102: In the reference model storage unit 12, plural reference models are collected and registered in advance. These reference models are prepared in advance for representative shape designs. In addition, the reference model is expanded at any time by collecting the design information of the past shape design from the simulation device 1 or an external server. In addition, the user can also collect reference models according to needs and register them in the reference model storage unit 12 via the reference model registration unit 42.

In these reference models, the design data that can be compared with the design data of the reference model are included together, and it also includes performance data and indicators representing its evaluation (hereinafter referred to as "performance indicators") ). Such performance indicators are prepared in advance through simulation processing or actual measurement experiments.

Step S103: The user registers the analysis parameters used for the analysis processing to the analysis processing unit 20 via the parameter registration unit 43. The analysis parameters here are, for example, various conditions of the analysis process or the number of iterations of the analysis process or the total number of selected reference models.

Step S104: The model setting unit 21 selects the reference model and the reference model used in the analysis process based on the parameters related to the model selection among the analysis parameters. As a reference model, it is possible to randomly select plural numbers from the registered reference model group, or to set plural reference models that have a certain degree of commonality or similarity with the reference data in the shape design data .

Step S105: Based on FIG. 3, the operation of the space selection unit 22 will be described. First, the space selection unit 22 formulates the divisional regions of the reference model 100 which are divided into plural regions based on the parameters (number or size, etc.) related to the divisional regions among the analysis parameters. The space selection unit 22 selects a space (hereinafter, referred to as a "division space 101") in which the reference model is divided so as to include these division areas.

The division space 101 can be a plane or a curved surface caused by a polygon, a circle, an ellipse, or a combination of these, or a three-dimensional columnar or polyhedron, a sphere or an ellipsoid, or a combination of these The resulting 3-dimensional space. The space selection unit 22 cuts out the design data group (division area) from the reference model 100 by dividing the space 101.

For example, when the driving impedance of the reference model 100 is reduced, the design data group corresponding to "the surface of the object that affects the driving impedance" is cut out by the division space 101. Also, for example, when the weight reduction and strength of the reference model 100 are improved in a good balance, the design data group inside the object that will affect the weight reduction and strength performance is obtained by dividing the space 101 Was cut out.

Furthermore, the space selection unit 22 obtains the division space 151 by respectively mapping and converting the division space 101 relative to the reference model 100 to the reference model 150. This mapping conversion is based on the reference points R and R'based on the alignment between the reference model 100 and the reference model 150 to convert the division space 151 of both sides into the corresponding position and size of the division space 101 Way, and is set. The space selection part 22 cuts out the design data group (corresponding area) of the reference model 150 by dividing the space 151.

Step S106: The analysis index extraction unit 23 extracts the difference in design data (the difference in shape or physical characteristics) between the distinguished area of the reference model and the corresponding area of the reference model for each of the reference models. The analysis index extraction unit 23 converts the difference index into an analysis index.

For example, in the case shown in FIG. 3, the analysis index extraction unit 23 is designed to compare the driving impedance and use the distance between the point coordinates as the difference to extract the design data of the point coordinates constituting the surface of the object. , And use the representative value of these differences (accumulated value or average value, central value or mode, etc.) as the analysis index.

Figure 4 is an explanatory diagram showing the difference histogram between the distinguished area of the reference model and the corresponding area of the reference model. As shown in the figure, since the distinguished area and the corresponding area are relatively small, the unevenness of the difference within the area converges within a meaningful width. Therefore, by using the representative value of the difference as an analysis index, the representative value can be used to index the tendency of the difference between regions.

Step S107: The analysis unit 24 performs correlation analysis and regression analysis between the analysis index and the performance index for each of the divided regions of the reference model. Furthermore, when the analysis index is a plural number, the analysis unit 24 performs multiple regression analysis between the plural analysis index and the performance index.

Fig. 5 and Fig. 6 are diagrams for explaining the relationship between the analysis index and the performance index for the divided regions 110-112 of the reference model.

In the distinguished area 110, the analysis index and the performance index show a negative correlation, and there is a tendency that the performance index will increase as the analysis index decreases in the negative direction (decreasing the height of the surface shape, etc.). Also, in the division area 111, the correlation between the analysis index and the performance index is low. On the other hand, in the distinguished area 112, the analysis index and the performance index show a positive correlation, and there is a tendency for the performance index to increase as the analysis index increases in the positive direction (increasing the height of the surface shape).

Step S108: The analysis unit 24 is aimed at "whether the analysis has been completed for the appropriate number of divided regions at appropriate intervals for the reference model" and "whether the analysis result has been obtained that exceeds the correlation coefficient required in the correction of the reference model" And so on, and make a judgment. When this determination condition is not satisfied, the analysis unit 24 returns the action to step S105, and issues an instruction to the space selection unit 22 to further select the division space. In this way, the actions of steps S105 to S108 are repeated until a specific condition is satisfied (or until it ends due to an interruption). On the other hand, if the determination condition is satisfied, the analysis unit 24 shifts the operation to step S109.

Step S109: The analysis result display unit 26 is based on evaluation judgments such as "the correlation coefficient is high", "the regression coefficient is large", "there is an analysis index with a large multiple regression coefficient", and "the amount of shape change is large". Visualize the distinguished areas where the benchmark model has a high impact on the aforementioned performance indicators. For example, the above-mentioned distinguished areas that have an impact on performance improvement are displayed in the shape of a heat map with color added, and are displayed as a composite display on the 3D display of the reference model.

In addition, for example, by performing the information display as shown in FIG. 6, it can also be combined with the display that the areas 110 and 112 are parts that affect the performance improvement, and provide an image to the user It is information such as "what kind of correlation is the direction in which design changes should be made".

<Design example caused by simulation device 1> Next, the shape design of the elevator car using the simulation device 1 will be described. Fig. 7 is an analytical model of the elevator car, and representatively shows the reference model 200a, reference models 200b, and 200c.

In this figure, 201a to 201c are shown on top of the reference model 200a, reference models 200b, and 200c. 202a-202c are displayed on the sides of the reference model 200a, reference models 200b, and 200c.

Here, in order to reduce the traveling resistance when the elevator is ascending, the surface shape 203a of the reference model 200a on the elevator is set to improve the aerodynamic performance.

Here, with respect to the surface shape 203a, the surface shapes 203b and 203c of the reference model are used as comparison objects. At this time, since each model has a symmetrical shape centered on the reference points 205a, 205b, and 205c, the target area is limited to 206a, 206b, and 206c. Next, in the target regions 206a, 206b, and 206c, the ellipsoidal zoning spaces 204a, 204b, and 204c are selected as arbitrary zoning spaces, and the center position of the ellipsoid and the size of the ellipse shape are used as parameters.

Fig. 8 shows a state where the surface shape 203a of the reference model 200a and the surface shape 203b of the reference model 200b are overlapped as the cross-sectional shape of the elevator top. Here, when an arbitrary ellipsoid division space 210 is selected, the divisional area of the surface shape 203a cut out by the division space 210 and the corresponding area of the corresponding surface shape 203b are The difference value of the height between them is extracted as a histogram 211.

Calculate the average value for this histogram 211 and select it as the analysis index of the reference model 200b. The distribution map (hereinafter referred to as the "correlation distribution") between this analysis index and the performance index of the reference model 200b is used as the drawing point 212 And be expressed.

By performing the same description for most other reference models, the correlation distribution between the analysis index and the performance index is obtained. As a result of the analysis, if the correlation is negative, the reference model 200a can be corrected by moving the partitioned space 210 in the direction of decreasing the height of the surface shape 203a, so that the running resistance can be reduced. If it is a positive correlation, the reference model 200a can be corrected by facing the direction of increasing the height of the surface shape 203a in the opposite direction to the partition space 210, so that the running resistance can be reduced.

<Effects of Example 1> (1) As mentioned above, in Example 1, if the distinguished area of the reference model is an area that will have an impact on the performance improvement, then in the distinguished area, there is a relationship between the analysis index and the performance index. Specific tendency (related). On the contrary, if the distinguished area of the reference model is an area that is not related to performance improvement, then there will be no specific tendency (correlation) between the analysis index and the performance index in the distinguished area. Therefore, by obtaining the correlation between the analysis index and the performance index, it becomes possible to determine whether the distinguished area of the reference model will have an impact on the performance improvement, and then to determine the extent of the impact.

(2) In addition, in the technique of Patent Document 1, there is a problem that it is difficult to narrow the range to an important shape that affects performance. However, in the first embodiment, since it is possible to determine whether it is an area that affects the performance improvement for the distinguished area of the reference model, it is important to be able to appropriately narrow the range to affect the performance. place.

(3) Furthermore, in the technique of Patent Document 1, there is a problem that it is difficult to implement a complex shape that cannot select design parameters that affect performance. However, in Example 1, by using the difference in shape and the difference in physical properties as the analysis indexes, it becomes possible to obtain the correlation between the performance index and the analysis index, so as to reduce the impact on performance. The parameter factors are the shape or the physical characteristics."

(4) Furthermore, in the technique of Patent Document 1, since the calculation load in the first performance analysis is large, when the divided parts are changed only slightly at a time, the huge number of In the case of combining the morphology to perform densely repeated performance analysis, there is a problem that it takes an extremely long time until the optimal solution is obtained. However, in Example 1, the performance index of the reference model can be obtained independently of the reference model. Therefore, since the performance index of the reference model only needs to be obtained in advance, the calculation amount of the analysis unit 24 itself is small, and the analysis result of the reference model can be obtained in a short time.

(5) Furthermore, in Embodiment 1, following the specific tendency (correlation) shown by the analysis index and the performance index, it can be determined that if the analysis index is moved in which direction, the performance index will be improved.

(6) Furthermore, in Embodiment 1, the influence of the reference model on the performance index at the distinguished area is displayed on the analysis result display unit 26 in response to the correlation obtained for the distinguished area. Therefore, it is possible to notify the user of the distinguished area that will become an important part in the performance improvement of the reference model, and it becomes possible to appropriately support the user's design work.

(7) Furthermore, in Example 1, the segmented area evaluated as having a high correlation coefficient is specifically visualized by the analysis result display unit 26. Therefore, it is possible to let the user know the distinguished area that will surely respond to the change of the analysis index and change the performance index.

(8) In addition, in Example 1, the classification area evaluated as a large regression coefficient is visualized by the analysis result display unit 26. Therefore, it is possible to let the user know the distinguished areas that greatly change the performance index relative to the change of the analysis index.

(9) Furthermore, in Example 1, the analysis result display unit 26 visualizes the distinguished area evaluated as having an analysis index with a large multiple regression coefficient. Therefore, it is possible for the user to know the analysis index that will affect the performance improvement among most of the analysis indexes such as the difference of shape or the difference of various physical characteristics.

(10) Also, in Example 1, the segmented area evaluated as a large amount of shape change is visualized by the analysis result display unit 26. The so-called shape change is a large division area, which is a large division area between a plurality of reference models and the difference in shape (analysis index). Such a distinguished area is a place that needs attention, although the influence on the performance index is large, but the shape change is likely to increase when the reference model is corrected. The system can let users know the places that need attention.

(11) Furthermore, in the first embodiment, it is also possible to use a partitioned space composed of polygons, circles, ellipses, cylinders, polyhedrons, spheres, ellipsoids, and combinations of these including the partitioned regions of the reference model. . Therefore, even if it is a divided area of various shapes, it can be freely matched by cutting out the diversified space.

(12) Furthermore, in Embodiment 1, the corresponding area of the reference model is determined by transforming the divisional space map including the divided areas of the reference model to the reference model. If there is a certain degree of positional deviation between the reference model and the reference model, then the corresponding area will be slightly offset from the reference model if only the division area of the reference model is mapped to the reference model. The setting of the corresponding area becomes difficult. However, by using the division space that surrounds the division area, even if the position between the reference model and the reference model is offset, it becomes possible to set the corresponding area by enclosing the division space. Therefore, by using the division space, the number of reference models that can be used in the analysis of the reference model can be increased.

(13) Furthermore, in Example 1, as an analysis index, it is possible to use the difference in shape or physical characteristics between the distinguished area and the corresponding area. If the difference in shape is used as an analysis index, it becomes possible to support the user for the correction of the shape (height, inclination, step, curvature, unevenness, etc.) of the reference model. On the other hand, if the difference in physical properties is used as an analysis index, it becomes possible to support the user in correcting the physical properties (material, surface roughness, strength, elasticity, density, weight, etc.) of the reference model.

(14) In addition, in Embodiment 1, it is possible to obtain representative values such as integrated value, average value, mode, median value, etc., for the difference between the spatially sampled complex numbers for the distinguished area and the corresponding area. , And used as an analysis indicator. By using the representative value in the analysis index in this way, it becomes possible to grasp the correlation between the analysis index and the performance index statistically (good and rough). Therefore, the system of providing information related to the correction of the reference model does not become excessively detailed, and can support practical corrections. [Example 2]

Next, as the second embodiment, the simulation device 1A that automates the correction of the reference model will be described.

<Configuration of Example 2> FIG. 9 is a diagram showing the structure of the simulation device 1A. In this figure, the same reference numerals are assigned to the same configuration as in the first embodiment (refer to FIG. 1), and the repeated description here is omitted.

The simulation device 1A is provided in the analysis processing unit 20, and includes a model correction unit 27, a correction model display unit 28, and a model analysis unit 29. The model correction unit 27 automatically corrects the reference model in the direction of the analysis index that will increase the performance index. The corrected model display unit 28 displays the corrected reference model. The model analysis unit 29 analyzes the performance index for the revised reference model.

<Action of Example 2> Fig. 10 is a flowchart illustrating the operation of the simulation device 1A. In this figure, since steps S101 to S109 are the same operations as in the first embodiment, the repeated description here is omitted. Hereinafter, the operation after step S110 will be described.

Step S110: The model modification unit 27 selects a plurality of distinguished regions that are evaluated as having a high influence on the performance index based on the analysis processing of the analysis unit 24. For each of the selected divisional areas, the model correction unit 27 temporarily corrects the reference model locally in the direction of the analysis index that will increase the performance index. The model correction unit 27 comprehensively corrects the reference model based on these local temporary corrections.

This kind of comprehensive correction is carried out by processing the boundary line or gap between the divided areas to make the changes of the connection shape continuous with each other for the divided areas that become the object. In addition, the comprehensive correction can also be used for smoothing the local temporary correction on the entire reference model. Furthermore, the comprehensive correction can also be performed by processing (curved surface regression analysis) that approximates the surface or plane of the area that has been temporarily corrected locally on the 3-dimensional space of the reference model. It.

Step S111: The model analysis unit 29 performs analysis processing on the corrected reference model and calculates a performance index. Here, in the analysis process, various analysis processes such as fluid analysis, strength analysis, or weight analysis can be used.

Step S112: The reference model storage unit 12 groups the revised reference model and performance index as a group, and registers it as a new reference model, thereby expanding the reference model database.

Step S113: Based on the analysis parameters set in the model setting unit 21, the model analysis unit 29 determines whether the performance index of the revised reference model meets a specific reference.

When the specific criterion is not satisfied, the operation is returned to step S110, and the model correction unit 27 re-corrects the reference model by adjusting the correction amount of the reference model.

In addition, in the case where the specific criteria etc. are not satisfied even after repeated corrections, the operation is returned to step S104, and the setting of the analysis parameters (position and size of the division area, etc.) is changed, and steps S104~ The operation of step S110 is performed again. In this case, the department can also use the revised reference model as a new reference model.

On the other hand, when the specific criterion is satisfied, the model correction unit 27 shifts the operation to step S114.

Step S114: The corrected model display unit 28 displays the corrected standard model that satisfies the specified standard.

<Effects of Example 2> Example 2 exhibits the following effects in addition to the effects of Example 1.

(1) In the second embodiment, the reference model is corrected in the direction of the analysis index that makes the performance index higher, so that automatic correction can be performed to improve the performance of the reference model.

(2) Furthermore, in the second embodiment, the localized temporary correction is preferentially performed for the plural divided areas evaluated as having a high influence on the performance index. Furthermore, the divisional areas other than that are used as free areas when comprehensively correcting the reference model. Therefore, it becomes possible to perform automatic corrections that give priority to improving the performance of the reference model locally.

(3) In the second embodiment, when the corrected reference model does not meet the specific reference, the correction amount of the reference model is adjusted to perform the re-correction of the reference model. In this case, it becomes possible to perform re-correction of the reference model by light processing that only adjusts the correction amount.

(4) Furthermore, in the second embodiment, when the revised reference model does not meet a specific reference, the setting of the analysis parameters is changed or the reference model is replaced with "the revised reference Model” and so on, to repeatedly perform automatic correction of the reference model. In this case, it becomes possible to re-correct the reference model more fundamentally.

(5) Furthermore, in the second embodiment, the revised reference model is memorized as a reference model including the performance index analyzed by the model analysis unit 29. Therefore, it becomes possible to automatically expand the number of reference models memorized in the reference model memory unit 12.

〈Supplementary matters of implementation form〉 In the above-mentioned embodiment, although the design purpose of the shape of the elevator car is explained, the purpose of the present invention is not limited to this. For example, the system can also be applied to the shape design of the nose shape of an airplane, the shape design of the body shape of an automobile, the shape design of the nose shape of a railway vehicle, or other purposes.

In addition, the present invention is not limited to the above-mentioned embodiments, but includes various modifications. For example, the above-mentioned embodiments are described in detail for the purpose of explaining the present invention easily, and the present invention is not limited to those that include all the structures described above. In addition, a part of the configuration of a certain embodiment may be replaced with a configuration of another embodiment, and the configuration of another embodiment may be added to the configuration of a certain embodiment. Furthermore, it is possible to perform addition, deletion, and replacement of other configurations for a part of the configuration of each embodiment.

1: Analog device 1A: Analog device 10: Model Management Department 11: Reference model memory 12: Reference model memory 20: Analysis Processing Department 21: Model Setting Department 22: Space Selection Department 23: Analysis index extraction department 24: Analysis Department 26: Analysis result display section 27: Model Correction Department 28: Modified model display 29: Model Analysis Department 40: Operation Department 41: Reference Model Registration Department 42: Reference Model Registration Department 43: Parameter Registration Department 100: Benchmark model 101: Zoning Space 110: Distinguish area 150: Reference model 151: Zoning Space

[Fig. 1] Fig. 1 is a diagram showing the structure of the simulation device 1. [FIG. 2] FIG. 2 is a flowchart for explaining the operation of the simulation device 1. [Fig. 3] Fig. 3 is a diagram for explaining the operation of the space selection unit 22. [Figure 4] Figure 4 is a histogram showing the difference between the distinguished area and the corresponding area. [Figure 5] Figure 5 is a diagram illustrating the relationship between analysis indicators and performance indicators. [Figure 6] Figure 6 is a diagram illustrating the relationship between analysis indicators and performance indicators. [Figure 7] Figure 7 is a diagram showing the analytical model of the elevator car. [Fig. 8] Fig. 8 is a diagram for explaining the analysis procedure of the simulation device 1. [Fig. 9] Fig. 9 is a diagram showing the structure of the simulation device 1A. [Fig. 10] Fig. 10 is a flowchart for explaining the operation of the simulation device 1A.

1A: Analog device

10: Model Management Department

11: Reference model memory

12: Reference model memory

20: Analysis Processing Department

21: Model Setting Department

22: Space Selection Department

23: Analysis index extraction department

24: Analysis Department

26: Analysis result display section

27: Model Correction Department

28: Modified model display

29: Model Analysis Department

40: Operation Department

41: Reference Model Registration Department

42: Reference Model Registration Department

43: Parameter Registration Department

Claims (12)

A simulation device, characterized in that it has: The reference model memory part is used as the processing object and memorizes the reference model; and The reference model memory is used as the comparison target of the aforementioned reference model, and the reference model with performance indicators is used as a plural memory; and Analysis processing department, The aforementioned analysis processing unit is equipped with: The space selection unit sets the corresponding area of each of the aforementioned reference models for the division areas of the aforementioned reference models; and The analysis index extraction unit is for each of the aforementioned reference models, and respectively extracts the difference between the aforementioned division area of the aforementioned reference model and the aforementioned corresponding area of the aforementioned reference model as an analysis index; and The analysis part is based on the aforementioned division area of the aforementioned reference model, and for the aforementioned analysis index obtained for each of the aforementioned reference models and the aforementioned performance index of each of the aforementioned reference models, to obtain correlation . Such as the analog device described in item 1 of the scope of patent application, in which: The aforementioned analysis processing unit is equipped with: The analysis result display unit displays the influence of the benchmark model on the aforementioned performance indicators in the aforementioned segmented area in response to the aforementioned correlation obtained by the aforementioned analysis unit for the aforementioned segmented area. Such as the analog device described in item 2 of the scope of patent application, in which: The aforementioned analysis result display part is based on the aforementioned correlation, and is based on the "correlation coefficient is high", "regression coefficient is large", "there are the aforementioned analysis indicators with a large multiple regression coefficient", and "shape change is large" At least one of the evaluation judgments is to visualize the aforementioned distinguished area where the aforementioned reference model has a high influence on the aforementioned performance index. Such as the simulation device described in any one of items 1 to 3 of the scope of patent application, wherein: The aforementioned space selection unit formulates a divisional space including polygons, circles, ellipses, cylinders, polyhedrons, spheres, ellipsoids, combinations of these, etc., of the aforementioned divided regions of the aforementioned reference model, and targets the aforementioned reference models For each of them, the aforementioned divisional space is mapped and converted to set the aforementioned corresponding area. Such as the simulation device described in any one of items 1 to 3 of the scope of patent application, wherein: The aforementioned analysis index extraction part is used as the aforementioned analysis index to obtain the difference in shape or physical characteristics. Such as the simulation device described in any one of items 1 to 3 of the scope of patent application, wherein: The aforementioned analysis index extraction unit is based on the difference between the spatially sampled complex numbers in the aforementioned division area and the aforementioned corresponding area, and also extracts representative values of integrated value, average value, mode, median value, etc., as The aforementioned analysis indicators. Such as the simulation device described in any one of items 1 to 3 of the scope of patent application, wherein: The analysis processing unit includes a model correction unit that corrects the reference model toward the analysis index where the performance index will increase. Such as the analog device described in item 7 of the scope of patent application, in which, The aforementioned model correction part is for each of the aforementioned divided regions which are evaluated as having a high influence on the aforementioned performance index, and the aforementioned reference model is oriented to the direction of the aforementioned analysis index that will increase the aforementioned performance index. Temporarily make corrections, and comprehensively make corrections to the aforementioned reference model based on the partial temporary corrections of the plural of the aforementioned divided regions. Such as the simulation device described in item 8 of the scope of patent application, in which: The aforementioned analysis processing unit is equipped with: The model analysis section analyzes the aforementioned performance indicators for the aforementioned reference model that has been modified. The model modification unit repeatedly performs modification of the reference model in order to make the performance index analyzed by the model analysis unit meet a specific standard. Such as the analog device described in item 9 of the scope of patent application, in which: The reference model memory unit increases the number of reference models by storing the revised reference model as the reference model having the performance index analyzed by the model analysis unit. . A simulation program characterized by: The information processing device functions as the reference model storage unit, the reference model storage unit, and the analysis processing unit described in any one of items 1 to 10 in the scope of the patent application. A simulation method characterized by: The step of obtaining the reference model is to obtain the reference model as the processing object; and The reference model acquisition step is to use the reference model with performance indicators as the comparison target of the aforementioned reference model, and obtain the plural number of the reference model; and Analysis processing steps, The aforementioned analytical processing steps include: The space selection step is to set the corresponding area of each of the aforementioned reference models for the distinguished areas of the aforementioned reference models; and The analysis index extraction step is for each of the aforementioned reference models, and the difference between the aforementioned distinguished area of the aforementioned reference model and the aforementioned corresponding area of the aforementioned reference model is obtained as an analysis index respectively; and The analysis step is based on the aforementioned distinguished area of the aforementioned reference model, and the aforementioned analysis index obtained for each of the aforementioned reference models, and the aforementioned performance index of each of the aforementioned reference models, to obtain the correlation .

Images