WO2006046737A1 - 多変数モデル解析システム、方法、プログラム、およびプログラム媒体 - Google Patents
多変数モデル解析システム、方法、プログラム、およびプログラム媒体 Download PDFInfo
- Publication number
- WO2006046737A1 WO2006046737A1 PCT/JP2005/019960 JP2005019960W WO2006046737A1 WO 2006046737 A1 WO2006046737 A1 WO 2006046737A1 JP 2005019960 W JP2005019960 W JP 2005019960W WO 2006046737 A1 WO2006046737 A1 WO 2006046737A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- model
- variables
- models
- cluster
- variable
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/02—CAD in a network environment, e.g. collaborative CAD or distributed simulation
Definitions
- Multivariable model angle analysis system method, program, and program medium
- the present invention provides a multivariable model analysis system, method, program, and program medium for extracting design principles such as basic causal relationships inherent in an analysis model by analyzing the relationship between a plurality of variables and characteristic values. About.
- the “design principle” here refers to the action, characteristics, basic causal relationship between each factor, etc. inherent in the design model. For example, in a vehicle suspension structure, there is a physical causal relationship between the design variables such as the coordinates of each link and the characteristic values of the camber angle and toe angle. What is simplified so that it is easy to understand is called a design variable.
- this method merely makes it possible to reduce design variables by simply referring to the sensitivity of each part, and does not extract the design principles inherent in the model to be designed. For this reason, it is impossible to extract a design principle that connects a design variable and a characteristic value, and it is difficult for a designer to grasp a design policy.
- the present invention has been made in view of the above-mentioned problems, and the problem to be solved by the present invention is that a high-efficiency simulation and design can be performed by extracting design principles in a multivariable analysis model.
- a multivariable model analysis system, method, program, and program medium
- the present invention calculates a characteristic value of the model based on a model generation unit that generates a plurality of models each having a plurality of design variables, a given model variable, A characteristic value calculation unit that writes variables and characteristic values of the model, a clustering unit that classifies a plurality of models approximated by the characteristic values into the same cluster, and a correlation between model design variables in each cluster A correlation coefficient calculation unit that calculates the number and writes the correlation coefficient in the memory map, and an extraction unit that extracts design variables whose correlation coefficient exceeds a predetermined value in each cluster from the fir memory map.
- the model generation unit may determine a plurality of design variables using an orthogonal table. Generate a model with
- the clustering unit clusters models having the smallest distance between characteristic values into the same cluster.
- the correlation coefficient calculation unit calculates a correlation coefficient between the design variables by changing a variable related to the design variable.
- the extraction unit calculates an average value of correlation coefficients of the design variables in a plurality of 'clusters, and extracts design variables whose average value exceeds a predetermined value from the memory map.
- the database further includes a database for storing and retrieving the design variables extracted by the extraction unit.
- a model generation unit that generates a number of models each having a plurality of variables, a characteristic value of the model based on the given model variables, and the model A characteristic value calculation unit that writes the variables and characteristic values of the above and a plurality of models having high similarity in the characteristic values are classified into the same cluster to generate a cluster group, and a space on the space having the characteristic values as coordinate axes.
- the cluster group is arranged at the same time, and a clustering unit that sequentially samples clusters located on a straight line, a curve, or a plane showing a desired change in the characteristic value of the cluster group, and a sampling order by the clustering unit And a principle extraction unit that determines how the average value of the model variables included in each sampled cluster changes. That.
- the model generation unit generates a plurality of orthogonal tables by changing an array of a plurality of factors assigned to the orthogonal tables, and generates a plurality of models using these orthogonal tables.
- classification is performed based on the values of design variables that have a large influence on the characteristic values. . That is, there is little influence on the characteristic value Sensitivity to design variables can be reduced, and design variables that are strongly associated with characteristic values can be identified.
- FIG. 1 is an overall view of an analysis system according to the present invention.
- FIG. 2 is a hardware block diagram of the analysis system according to the present invention.
- FIG. 3 is a functional block diagram of the analysis system according to the present invention.
- FIG. 4 is a flowchart showing the operation of the analysis system according to the present invention.
- FIG. 5 is a flowchart showing the operation of the analysis system according to the present invention.
- FIG. 6 is a conceptual diagram of an analysis system according to the present invention. '
- FIG. 7 is a perspective view of a double wishbone suspension according to the first embodiment of the present invention.
- FIG. 8 is a diagram showing the joint points of the double wishbone suspension according to the first embodiment of the present invention.
- FIG. 9 is a diagram for explaining design variables and characteristic values according to the first embodiment of the present invention.
- FIGS. 1 OA and 1 OB are diagrams for explaining a camber angle and a toe angle according to the first embodiment of the present invention.
- FIG. 11 is a diagram showing hierarchical clustering according to the first embodiment of the present invention.
- FIG. 12 is a graph of the cluster corners and corners of cluster 1 according to the first embodiment of the present invention.
- FIG. 13 is a graph of the camber angle and the triangle of the cluster 6 according to the first embodiment of the present invention.
- FIG. 14 is a graph of the camper angle and the triangular angle of cluster 7 according to the first embodiment of the present invention.
- FIG. 15 is a diagram showing the correlation between the “ff” variables in cluster 1 according to the first embodiment of the present invention.
- FIG. 16 is a diagram showing the correlation between the variables in the cluster 6 according to the first example of the present invention.
- FIG. 17 is a diagram showing the correlation between the descriptive variables in the cluster 7 according to the first embodiment of the present invention.
- FIG. 18 is a diagram showing joint points having a strong correlation in the double wish suspension according to the first embodiment of the present invention.
- FIG. 19 is a diagram showing an average value of correlation coefficients according to the first embodiment of the present invention.
- FIGS. 20A, 20B, and 20C are diagrams for explaining the design principle of the Dadar wishbone suspension according to the first embodiment of the present invention.
- FIG. 21 is a graph showing a comparison between the analysis result and the actual measurement value of the camber angle characteristic according to the first example of the present invention.
- Figure 22 shows the analysis results and actual results for the toe angle characteristics according to the first embodiment of the present invention. It is a graph showing the comparison with a measured value.
- FIG. 23 is an example applied to a semiconductor integrated circuit according to the second embodiment of the present invention.
- FIG. 24 is a graph for explaining the delay time of the inverter circuit according to the second embodiment of the present invention.
- FIG. 25 is a block diagram of a MOS transistor according to the second embodiment of the present invention.
- FIG. 26 is a diagram for explaining an orthogonal table according to the present invention.
- FIG. 27 is a flowchart showing the operation of the analysis system according to the present invention.
- FIG. 28 is a diagram for explaining the cluster sampling method according to the present invention.
- FIG. 29 ' is a graph showing the average value of the design variables of each cluster according to the present invention.
- FIG. 3 shows a model of the multilink suspension according to the fourth embodiment of the present invention.
- FIG. 31 is a graph showing the distribution of design variables according to the fourth example of the present invention.
- FIG. 32 is a graph showing clustered design variables according to the fourth embodiment of the present invention.
- FIG. 33 is a diagram for explaining hierarchical clustering according to the fourth embodiment of the present invention.
- FIG. 34 is a specific example of design variables according to the fourth embodiment of the present invention.
- FIG. 35 is a diagram showing the average values of the design variables according to the fourth example of the present invention.
- FIG. 36 is a view for explaining cluster sampling according to the fourth embodiment of the present invention.
- FIG. 37 is a graph showing the tendency of changes in design variables according to the fourth example of the present invention.
- FIG. 38 illustrates the sampling of clusters according to the fourth embodiment of the present invention.
- FIG. 39 is a graph showing the trend of changes in design variables according to the fourth example of the present invention.
- FIG. 40 is a diagram for explaining cluster sampling according to the fourth embodiment of the present invention.
- FIG. 41 is a graph for explaining the tendency of changes in design variables according to the fourth embodiment of the present invention.
- Fig. 1 shows the overall configuration of the analysis system according to this embodiment.
- This analysis system includes a simulation terminal 1, an emulator terminal 2, a hardware interface 3, a server 4, and a network 5.
- the simulation terminal 1 includes a computer main body, a display, a keyboard, and the like, and has a function of simulating an analysis model, clustering design numbers, and extracting design variables.
- the simulation terminal 1 can extract the design principle inherent in the analysis model based on the extracted design variables and store this information as a database. As a result, it is possible to search for desired design information according to a request from the user.
- the emulator terminal 2 is equipped with a hard air interface 3 for measuring physical quantities of the prototype hardware. This emulator terminal 2 is used to perform detailed design by executing emulation by HIL (Hardware In Loop) based on the design variables calculated by the simulation terminal 1.
- the server 4 is connected to the simulation terminal 1 and the emulation terminal 2 via a network 5 such as a LAN or WAN, and performs a predetermined calculation process in response to requests from these terminals 1 and 2. It is for execution. In this embodiment, it is also possible to execute simulation processing, database search, etc. in place of the above-described terminals 1 and 2.
- the server 4 has a function of storing a simulation result such as a correlation coefficient and a model design principle calculated by the simulation terminal 1 as a database, and performing a search of the database in response to a request from the user. ing.
- FIG. 2 is a diagram of the Harde Air mouth of the simulation terminal 1 described above.
- This simulation terminal 1 includes a CPU (Central Processing Unit) 101, a path 102, a memory 103, a HDD (Hard Disk Drive) 104, an external storage device 105, an operation interface 106, a video controller 107, an input / output interface 109, It consists of NIC (Network Interface Card) 1 10, Display 108, etc.
- the CPU 101 extracts the design variable area of the given analysis model and executes simulation calculation of the design variable area.
- the path 102 is constituted by a data bus opaque path, and is used to exchange data between the CPU 1001 and a device such as a memory.
- the memory 103 is used as a work area for the CPU 101 to execute the analysis program, and the HDD 104 is used to store the program and a database of simulation results.
- the external storage device 105 is a storage device for reading / writing data from / to various recording media such as MO, CD, CD-R, CD-RW, DVD, DVD-R, and DVD-RW. In addition to storing the analysis program according to the present embodiment on this recording medium, it is possible to save simulation results and the like.
- the operation interface 106 is connected to input devices such as a keyboard and a mouse. Through these input devices, the user can specify an analysis model and give a database search instruction to the simulation terminal 1.
- the video controller 107 includes a graphic memory, a 3D graphic controller, and the like, and has a function of converting an analysis model, a graph of a simulation result, and the like into a video signal.
- the display 108 is composed of a CRT, a liquid crystal display, etc., and is used to display an image based on the video signal from the video controller.
- the input / output interface 1 0 9 is configured by a USB, a serial port, a normal port, and the like, and is used for connecting an external device such as a printer and the simulation terminal 1.
- the NIC 110 is a network adapter for connecting the simulation terminal 1 to Ethernet (registered trademark), the Internet, or the like. It is also possible to download the analysis program from the server 4 to the simulation terminal 1 via the NIC 110.
- FIG. 3 is a functional block diagram of the analysis system according to this embodiment.
- the theoretical formula input unit 11 is used to determine the theoretical formula of the model to be analyzed.
- the theoretical formula expresses various models subject to simulation of mechanical structures such as vehicles, electronic circuits, economic phenomena, etc.
- the relationship between multiple design variables and characteristic values (function values) is expressed in mathematical formulas. It is a representation.
- the model generation unit 12 is for generating various models by determining a plurality of design variable values in the theoretical formula input by the theoretical formula input unit 11 1.
- an orthogonal table is used to determine uniform and unbiased design variables.
- This orthogonal table is composed of a table in which the levels of factors (design variables) are evenly allocated.
- an orthogonal table with 4 factors and 2 levels is composed of a table in which 1 and 2 are assigned evenly for each of the 4 factors, and each orthogonal value table is used to determine the values of the 4 design variables. Is possible. It is desirable to use an orthogonal table to generate such uniform design variables.
- design variables may be generated pseudo-equally using random numbers.
- each factor of a given orthogonal table a plurality of 3 ⁇ 4: intersection tables are generated, and the number of design variables to be sampled is increased by using the plurality of orthogonal tables thus obtained. It is possible.
- the simulation calculation unit 10 is for substituting each ft variable generated by the model generation unit 12 into a formula and calculating a characteristic value. For example, if 1 2 8 models are generated by the model generation unit 1 2, 1 2 8 data sets composed of characteristic values and design variable values are generated.
- the simulation calculation unit 10 is preferably capable of calculating with as high accuracy as possible, but can be configured using a general-purpose simulation system. As described above, since the number of design variables to be calculated is limited by the design variable generation unit 12, the calculation amount of the simulation calculation unit 10 can be minimized.
- the clustering unit 14 is used to classify models having approximate characteristic value changes using a hierarchical clustering technique.
- the multiple models there are multiple models whose characteristic value changes approximate each other, and these are classified into one cluster.
- the distance between the characteristic value of one model and the characteristic value of another model is calculated for each design variable, and the models having the smallest sum of these distances are classified into the same cluster.
- clustering should be performed using methods such as the group average method. Is also possible.
- the clustering unit 14 can generate a hierarchically classified cluster by sequentially classifying the classified clusters as higher-order clusters. ⁇
- the correlation coefficient calculator 13 calculates the correlation coefficient between design variables in each cluster. For example, when other variables or characteristic values related to the design variable are changed, the two design variables X and Y show a predetermined change.
- the presence or absence of correlation between changes in the two design variables can be expressed by the correlation coefficient of the following equation.
- x 0 and y O represent average values of x and y
- ⁇ x and S y represent standard deviations of x and y.
- the correlation coefficient r is 1 1 ⁇ r ⁇ 1. If the positive correlation is strong, r is close to 1. If the negative correlation is strong, r is close to 1. Show.
- the principle extraction unit 15 is for extracting the design principle inherent in the model based on each cluster.
- the principle extraction unit 15 The average of the coefficients is calculated, and the bonds of design variables that have a strong correlation in all clusters are obtained.
- These design variables work together to influence the characteristic values, and it is possible to grasp the design principles inherent in the model by diagramming the relationship between these design variables and the characteristic values. It becomes possible. For example, if three design variables of a number of design variables have a strong positive correlation, it can be seen that a desired characteristic value can be obtained by similarly changing these design variables. By modeling the relationship between these design variables and characteristic values, the user can understand the design principle of the model to be analyzed.
- Design variables can be automatically extracted by predetermining the correlation coefficient threshold.
- the design principle inherent in the model can be automatically extracted by calculating the relationship of the change between the extracted design variable and the characteristic value by the simulation calculation unit 10 or the like.
- the extraction process based on the ten principles may be performed by the user by displaying the correlation coefficient or the like on a display or the like.
- the output unit 18 is for visually presenting the above processing result to the user. For example, it is possible to display the classified clusters, the number of correlations between design variables in each cluster, and the relational expressions between the extracted design variables and characteristic values. Further, the output unit 18 may transmit the processing result to another terminal through the network.
- the database 17 stores the processing results as a database and enables retrieval. That is, by storing data such as classified clusters, correlation coefficients, and extracted design principles in the database 17 in advance, desired data can be searched from a large number of stored processing data, It can be used.
- Memory map 16 includes theoretical formula input unit 1 1, power generation unit 1 2, simulation calculation unit 1 0, correlation coefficient calculation unit 1 3, clustering unit 1 4, principle extraction This is for holding the processing result in each of the parts 15. Each processing unit can pass the processing result to another processing unit via the memory map 16.
- the analysis method shown in Fig. 4 is used to search for multiple design variables that affect the characteristic values of each other by calculating the correlation coefficient of each cluster, and to extract the design principles inherent in the model. It is a method.
- the user determines the theoretical formula of the model to be analyzed and inputs it to the theoretical formula input unit 11 (step S 4 0 1).
- the model to be analyzed is a mechanical structure, the three-dimensional coordinates of each part of the structure are represented as design variables, and the theoretical formula representing the strength, noise, etc. as characteristic values is determined.
- the model to be executed is a semiconductor device, it is possible to determine a theoretical formula that expresses the distance between wires, the electrode area, etc. as design variables, and the delay time as a characteristic value.
- economic phenomenon models it is also possible to extract the principles of economic phenomena by determining the theoretical equations between design variables and characteristic values.
- the model to be analyzed is not limited to a physical model, and may be various models such as an economic model.
- the model generation unit 12 specifically determines the values of the design variables in the determined theoretical formula, and generates a plurality of models (step S 4 0 2). In other words, the model generation unit 12 can generate a plurality of equal and non-overlapping models by defining each design variable using an orthogonal table. It should be noted that the number of generated models' is preferably a sufficient number to extract the design principle 1 2, for example 1 2 8. Stored in 6 The
- the simulation calculation unit 10 performs simulation of the generated model (step S 40 3). That is, the simulation calculation unit 1 ⁇ 0 substitutes the value of the specifically determined design variable into the theoretical formula, and calculates the characteristic value.
- the combination of characteristic values and design variables calculated in this way is stored in the memory map 16 for each model as a single data set.
- the clustering unit 14 calculates the distance between the characteristic values of each model (step S 40 4). For example, some 1 2 8 models have approximated changes in characteristic values, and the distance between such characteristic values is short.
- the clustering unit 1 ′ 4 classifies models having characteristic values with a short distance from each other into one cluster (step S 4 0 5). The changes in the characteristic values of the two models classified in this way are approximate to each other. Furthermore, the clustering unit 14 performs hierarchical clustering by classifying each cluster into a higher cluster.
- the correlation coefficient calculation unit 15 calculates the correlation coefficient between the design variables in the cluster obtained by the above processing (step S 4 0 6). That is, the correlation coefficient calculation unit 15 obtains the change of the design variable when the characteristic value or the like is changed within a certain range, and calculates the correlation coefficient between the design variables at this time.
- FIG. 6 shows an example of the strength of correlation between design variables.
- clusters 1 and 2 include models having design variables 1 to 8, and models classified into the same cluster have approximate characteristic value changes.
- design variables 1 to 8 those with strong correlation
- the figures are connected with solid lines, and those with weak correlation are shown with wavy lines. In this way, by designing the strength of correlation between design variables, it is possible to visually grasp the design variables that are linked to each other.
- the clustering unit 14 calculates an average value of the number of correlations of design variables in a plurality of clusters (step S 4 0 7).
- design variables 2, 5, and 8 have a strong correlation with each other, and it is confirmed that the correlation strength of these design variables is common to both clusters 1 and 2. it can. Therefore, even in the average values of the correlation coefficients of design variables in these clusters, the correlation coefficients of design variables 2, 5, and 8 show high values. In other words, the design variables 2, 5, and 8 have a strong correlation in common regardless of the cluster class, and the design variables 2, 5, and 8 are linked. It can be seen that the characteristic value is affected. '.
- the design principle of the model to be analyzed can be grasped visually (step S 4 0 8). Further, the principle extraction unit 15 extracts the design principle inherent in the model based on the relationship between the strongly correlated design variables and the characteristic values (Step S.40.09). For example, suppose that the model to be analyzed has outputs of design variables 1 to 5 and characteristic values 1 and 2, as shown in Fig. 5. Design variables 1 to 5 have a certain relationship with characteristic values 1 and 2, and the strong correlation is shown by thick arrows in the figure. A thin arrow indicates that the correlation is weak, and a variable that is not connected to the characteristic value by the arrow indicates that there is almost no correlation. Variables 2, 3, and 5 have a strong correlation with each other. By representing the connection between these variables and characteristic value 1 with arrows as shown in the figure, the design principle inherent in the model is visually represented. be able to.
- Database storage Data on models, clusters, design principles, etc. generated as described above are stored in the database 17. The user can search for desired data from the data stored in the database 17 and use it for design and the like.
- the analysis method shown in Fig. 27 is inherent in the model by sampling a plurality of clusters crossing a given plane or straight line in the cluster group distributed in space, and changing the design variables in the order of sampling. This method aims to extract the design principle.
- the user determines the theoretical formula of the model to be analyzed, and inputs it to the theoretical formula input section 11 (step 2 7 0 1).
- the analysis target model is not limited to a physical model such as a machine structure, but can be applied to various models such as a communication network and an economic model.
- the model generation unit 12 specifically determines the design variable values in the determined theoretical formula, and generates a plurality of models (step S 27 0 2). In other words, the model generation unit 12 can generate a plurality of models that are equal and have no overlap by defining each design variable using an orthogonal table. It is desirable that the number of generated models is sufficient to extract design principles, such as 1 28.
- the model generated in this way is stored in the memory map 16.
- a plurality of orthogonal tables may be generated by rotating one orthogonal table, and a model may be generated based on these orthogonal tables, or a model may be generated using random numbers.
- the simulation calculation unit 10 performs simulation of the generated model (step S 4 0 3). In other words, the simulation calculation unit 10 substitutes the value of the specifically determined design variable into the theoretical formula, and calculates the characteristic value. this The combination of the characteristic value calculated in this way and the design variable is stored in the memory map 16 as one data set.
- the clustering unit 14 calculates the distance between the characteristic values of each model (step S 40 4). For example, some 1 2 8 models have approximated changes in characteristic values, and the distance between such characteristic values is short.
- the clustering unit 14 classifies models having characteristic values with a short distance from each other into one cluster (step S 4 0 5). The characteristic value changes of the two models classified in this way are approximate to each other. Furthermore, the clustering unit 14 performs hierarchical clustering by classifying each cluster into a higher cluster.
- the clustering unit 14 calculates the average value of a plurality of design variables included in each cluster, and extracts the features of each cluster (step S 27 0 6). This average value is stored in memory map 16.
- the clustering unit 14 arranges clusters in the N-dimensional space using each of the N characteristic values as coordinate axes. For example, when there are three characteristic values, clusters are arranged in a three-dimensional space as shown in FIG. Then, the clustering unit 14 sequentially samples clusters that intersect the desired curve and curved surface with respect to the cluster group in the space (step S 27 07).
- the sampling curve and curved surface can be determined in consideration of which characteristic value is important. For example, in the cluster group in Fig. 28, the characteristic value corresponding to the Z-axis is greatly changed, the characteristic value corresponding to the X-axis is slightly changed, and the characteristic value corresponding to the Y-axis is made substantially constant. Let us assume a curve as shown by the arrows. The sampling unit 14 then selects the cluster intersecting this curve in the direction of the arrow. Sampling in order according to direction.
- sampling curve and phase may be determined automatically by the sampling unit 14 as well as being determined interactively by the operator looking at the clusters displayed on the display 10 8.
- sampling autopsy 14 can automatically detect envelopes and envelope surfaces surrounding a cluster group and sample clusters along these curves and curves.
- a straight line or plane parallel to any of the spatial coordinate axes may be automatically detected while traversing the cluster group.
- the principle extraction unit 15 determines how the design variable changes by changing the design variable of each sampled cluster and the sampling order of the clusters. For example, as shown in Figure 29, assume that the design variable C changes as shown by the arrow according to the sampling order of the clusters. In this case, it can be seen that changing the design variable C in the direction of the arrow changes the characteristic value as shown by the arrow ⁇ in Fig. 28. In other words, it is possible to extract a desired characteristic value and a design variable related to the characteristic value, and grasp how the desired characteristic value can be obtained by changing the design variable. It becomes possible to do.
- the graph and data obtained by the above processing are stored in the database 17 (step S 2 7 1 0).
- the user can search for desired data stored in the database 17 and use it for design.
- FIG. 7 is a conceptual diagram of a double wishbone suspension.
- the dub /! / Wishbone suspension includes a damper 70, an upper arm 7 1 and a mouthpiece arm 7 2 each having a substantially A shape.
- the upper arm 7 1 has joint points ⁇ 1, ⁇ 2, and ⁇ 3: ⁇ 3 ⁇ 4, and the joint point ⁇ 1 is connected to the kingpin side by a ball joint or the like.
- the joint points ⁇ 2 and ⁇ 3 are joined to the chassis side, and the upper arm 7 1 is configured to be freely movable around the axis connecting the joint points ⁇ 2 and ⁇ 3.
- the lower arm 7 2 also has joints ⁇ 4, ⁇ 5, and ⁇ 6 and constitutes a movable link with the upper arm 7 1. Even if the front / rear / left / right force is applied to the tire, only the vertical movement of the tire can be realized, and stable cornering performance can be obtained.
- the diamond is attached to the suspension with a predetermined inclination (camber angle) with respect to the contact surface (see Fig. 1 OA).
- a toe angle is provided so as to form a square shape with respect to the straight direction of the car (see Fig. 10B).
- camper angle and toe angle have a large influence on the driving performance of the car.
- Camber angle and toe angle are large depending on the position ⁇
- the design principle that is, the basic causal relationship existing between the coordinate values and the characteristic values of the joint points P1 to P7, which are design variables, can be obtained. It is possible to extract and easily understand the relationship between the two.
- the stroke of damper 70 is determined so that the Bump-Rebound range is ⁇ 9 O mm. .
- the model generation unit 12 determines each of the design variables P lx, Ply, ⁇ 1 ⁇ ... P7x, P7y, ⁇ 7 ⁇ using the orthogonal table.
- the design variables P lx, Ply, ⁇ 1 ⁇ ... P7x, P7y, ⁇ 7 ⁇ there are a total of 21 design variables, and 1 2 8 models are generated using an orthogonal table with 4 levels.
- a plurality of orthogonal tables may be derived by rotating one orthogonal table, and a model may be generated using these orthogonal tables.
- the simulation unit 10 calculates the camper angle and toe angle as characteristic values.
- the camber angle and toe angle for the design variables P lx, Ply, Plz-P7x, P7y, P7z defined in the orthogonal table are calculated, and these values are stored in a memory map as one data set. Stored in 1-6. In this way, the simulation is repeated for all 1 2 8 models.
- the characteristics of the camber angle and the toe angle differ depending on the model, but some 1 28 models have characteristics that are similar to each other.
- the clustering unit 14 obtains the distance between the curves representing the change in the camber angle and the toe angle for each of the 1 2 8 models, and classifies the models having the minimum distance into one cluster. In other words, models that approximate the characteristics of camber angle and toe angle are classified into one cluster.
- Figures 12 to 14 show examples of characteristics graphs of camber angles and toe angles classified into clusters.
- the horizontal axis represents the stroke amount (mm) of the damper 70
- the vertical axis represents the angle (deg).
- multiple characteristic curves drawn in each dull indicate that multiple models were classified into the same cluster.
- Clusters 1, 6, and 7 include the models shown in Figures 12 to 14 above.
- Figure 11 shows the results of hierarchical clustering according to this example.
- the clustering part 14 each of the 1 2 8 models is sequentially classified into a higher cluster. If layered clustering is continued, all of the 1 2 8 models are classified into one cluster, and the layered tree diagram shown in the figure is generated.
- the model was analyzed using a hierarchy having eight clusters 1-8. Each cluster contains a total of 16 models and is categorized based on design variables sensitive to camber and toe angle characteristics.
- the correlation coefficient calculation unit 13 calculates the phase coefficient of each design variable P 1 x, Ply, ⁇ 1 ⁇ ⁇ P7x, P7y, ⁇ 7 ⁇ . In other words, the correlation coefficient calculation unit 13 calculates the change in the design variables P lx, Ply, ⁇ 1 ⁇ ⁇ ⁇ 7 ⁇ , P7y, P7z when the damper 70 is changed within a certain range. Find the strength of the correlation between changes. ,
- Figures 15 to 17 show an example of the correlation coefficient calculation results for clusters 1, 6, and 7, respectively.
- the solid line indicates that the correlation coefficient is 0.95 to 1.00
- the dashed line indicates that the correlation coefficient is 0.90 to 0.95.
- the ten variables that are not connected by the solid line and the wavy line show that the correlation coefficient is below these values. This way (the diagram generated in this way is stored in database 17 and can be used for later design.
- the principle extraction unit 15 can design the design variables P l x, Ply, ⁇ 1 ⁇ of each cluster.
- the average correlation coefficient between P7x, P7y, and ⁇ 7 ⁇ is calculated, and the graph shown in Figure 19 Is generated.
- the solid line shows that the average value of the correlation coefficient is ⁇ 0.95 to 1.00.
- the combination of design variables P lz, P4z, and P 7z and the combination of design variables P 2x and P 3 have high correlation coefficients. ' ⁇
- the above design variables P lz, P4z, P 7z are shown in Fig.
- the design variables P lz, P4z, P 7z are the Z jobs at the junction points P l, P 4, P 7 From this, it can be seen that the displacement of the Z axis: ⁇ direction at the joint points P1, P4, and P7 has a significant effect on changes in the key bar angle and the toe angle. In other words, it can be seen that the camber angle and toe angle change as the joints P 1, P 4, and P 7 are displaced in the Z-axis direction due to the displacement of the damper 70.
- FIG. 20B is a view of the suspension viewed from the X-axis direction (vehicle traveling direction)
- FIG. 20C is a view of the suspension viewed from the Z-axis direction (vehicle upper direction).
- FIG. 20B when the damper 70 is moved up and down, the normal line changes as shown by the arrow, and the plane formed by the joint points PI, P4, and P 7 is displaced, It can be confirmed that the corner changes. Further, as shown in FIG. 20C, it can be confirmed that the surface formed by the joints PI, P4, and P7 is displaced and the toe angle is changed as the damper 70 moves up and down.
- Figure 21 shows the relationship between the camper angle and the displacement of the damper 70.
- the curve indicated by the black circle represents the camper angle change based on the analysis result
- the curve indicated by the black square represents the camber angle change calculated by the inclination of the normal line described above.
- Figure 23 shows an example of a semiconductor integrated circuit using CMOS transistors.
- This circuit includes two inverter circuits 2 3 1 and 2 3 2 and a wiring pattern 2 3 0 ′ connecting them.
- the C MO S transistor has gate capacitances C 3 and C 4 at the gate electrode and drain capacitances C 1 and C 2 at the drain electrode.
- the sizes of these gate capacitors C 3 and C 4 and drain capacitors C 1 and C 2 are determined by the structure of the gate electrode, the drain electrode, and the like.
- the wiring capacitance C 5 is determined by the distance between the wiring pattern 2 30 and other patterns. Depending on the size of these parasitic capacitances C1 to C5, the output waveform for the input waveform changes. For example, as shown in Fig.
- the output waveform has a delay time relative to the input waveform. That is, the output waveform delay time t dl with respect to the rising edge of the input waveform, and the output waveform delay with respect to the falling edge of the input waveform Time t d2 exists.
- the design principle inherent between the circuit structure and the delay times t dl and t d2 is extracted, and the delay times t dl and t d2 are efficiently obtained by designing. It can be reduced.
- the CMOS transistor gate length L, gate width W, gate oxide thickness d, and drain thickness P are defined as design variables in the CMOS transistor.
- the pattern length Ll and distance L2 between the wiring patterns 2 30 are defined as design variables. Then, delay times t dl and t d2 are determined as characteristic values, and a theoretical formula between them is determined and input to the theoretical formula input unit 1 1.
- the model generator 1 2 uses the orthogonal table to determine the design variable values for gate length L, gate width W, gate oxide thickness d, drain thickness P, pattern length Ll, and inter-wiring distance L2. Generate multiple models. As described above, by using the orthogonal table, it is possible to generate a model having uniform design variable values that do not overlap each other.
- the simulation unit 10 simulates these models and calculates the delay times t dl and t d2 of each model. The delay times t dl and t d2 calculated in this way show different changes depending on the value of the design variable. ⁇
- the clustering unit 14 approximates the change in the delay times t dl and t d2. Models are classified into the same cluster, and hierarchical clustering is performed. Each cluster is classified for each value of the design variable that has a large effect on the delay times t dl and t d2, and the design variables that have a large effect on the delay times t dl and t d2 are identified based on these clusters. can do. In this example, it can be confirmed that the design variables of the gate length L, the gate width W, the pattern length Ll, and the inter-wiring distance L2 have a relatively strong influence on the delay times t dl and t d2. .
- the correlation coefficient calculation unit 15 calculates the correlation coefficient between design variables in the cluster obtained by the above processing. That is, the correlation coefficient calculation unit 15 calculates a change in the design variable when the characteristic value or the like is changed within a certain range, and calculates the correlation coefficient between the design variables at this time. For example, when the gate capacitances C3 and C4 are changed, the correlation coefficient between the gate length L and the gate width W increases. In addition, when the wiring capacitance C5 is changed, the correlation coefficient between the pattern length Ll and the wiring distance L2 increases. In other words, design variables with high correlation coefficients interact with each other and affect the delay times t dl and t d2.
- the principle extraction unit 15 calculates an average value of correlation coefficients in a plurality of clusters, and searches for a design variable having a correlation coefficient having a high average value.
- the correlation coefficient of the gate length L and the gate ⁇ W and the correlation coefficient of the pattern length L l and the inter-wiring distance L 2 were also high.
- the gate capacities C 3 and C 4 are determined by the gate length L and the gate width W, and these values affect the delay times t dl and t d2.
- the wiring capacitance C 5 is determined by the pattern length L l and the inter-wiring distance L 2, and this value affects the delay times t dl and t d2.
- the present invention is also applicable to network line performance prediction. It is desirable that the network line 'has as large a capacity as possible. However, considering the maintenance cost of the line, a line having a larger size than necessary is not preferable in terms of cost. For these reasons, network performance is widely predicted.
- traffic XI, network configuration X2, number of users X3, number of applications X4, number of nodes X5 are assumed as network design variables, and variables X1 to X5 are determined by quantifying them.
- the network characteristic value is the Ping response as described above.
- the model generation unit 11 determines initial values of the variables X 1 to X 5 using the orthogonal table, and the simulation calculation unit 10 calculates the Ping response of the network model having this variable.
- the network model simulation can use the existing prediction system as it is.
- the clustering unit 14 classifies network models having characteristics that Ping responses are similar to each other into the same cluster. Furthermore, hierarchical clustering is performed by classifying these clusters.
- the correlation coefficient calculation unit 13 calculates design variables that are linked to each other by calculating a correlation coefficient between the design variables. For example, if the number of users increases, There is also a tendency for the number of applications to increase, and there is a strong correlation between the number of users X3 and the application X4 used. There is also a strong correlation between the network configuration XI and the number of nodes X5. Subsequently, the principle extraction unit 15 calculates the average value of the correlation coefficient in each cluster, and extracts design variables having a high correlation coefficient. In this way, design variables with strong correlations can be determined, and the relationship between these design variables and the characteristic value Ping response can be grasped. That is, according to the present embodiment, it is possible to extract the design principle between various design variables of the network model and the characteristic value Ping response.
- FIG. 30 shows a model of a multi-link suspension.
- This multi-link suspension includes a trailing arm, an inner arm, an upper arm, a mouth arm, and the like. Assume that the following analysis is performed using joint points P1, P2, P3, and P4 of each arm as design variables.
- the tire in order to improve the ground contact property and turning performance of the tire, the tire is attached to the suspension with a predetermined inclination (camber angle) with respect to the ground contact surface.
- a toe angle is provided so as to form a square shape in the straight running direction of the car.
- the theoretical formula of the multi-link suspension is input to the theoretical formula input unit 11.
- This theoretical formula can be used in advance for a simulation system or the like.
- This theoretical formula consists of the design variables P lx, Ply, ⁇ 1 ⁇ ⁇ ⁇ ⁇ ⁇ 4 ⁇ , P4y, ⁇ 4 ⁇ force consisting of the three-dimensional coordinate values of each of the junction points P 1 to P 4 defined as above.
- the characteristic value of the theoretical formula is the camber angle and the toe angle.
- the model generation unit 12 determines specific numerical values of the design variables P 1 x, Ply, ⁇ 1 ⁇ ⁇ P4x, P4y, and ⁇ 4 ⁇ using the orthogonal table.
- a plurality of orthogonal tables may be derived by rotating one orthogonal table, and a model may be generated using these orthogonal tables.
- the simulation unit 10 calculates the camber angle and toe angle as characteristic values.
- the characteristics of the camber angle and the toe angle differ depending on the model, but some 1 28 models have characteristics that are similar to each other.
- the clustering unit 14 obtains the distance between the curves representing the change in the camber angle and the toe angle for each of the 1 2 8 models, and classifies the models having the smallest distance into one cluster. In other words, a model approximating the characteristics of the camber angle Dells are classified into 1 ⁇ clusters.
- Figures 32 and 33 show the results of hierarchical clustering according to this example.
- the clustering unit 14 classifies each model into a higher cluster. If you continue to execute hierarchical clustering, all of the 1 2 8 models are classified into one cluster, and the hierarchical tree diagram shown in the figure is generated.
- the model was analyzed using a hierarchy having 17 clusters (1) to (; 17). Each cluster is classified based on design variables that are highly sensitive to camber and toe angle characteristics.
- Cluster 1 contains multiple design variables, as shown in Figure 32.
- Figure 33 shows the average values of these design variables. In this way, by plotting the average value of design variables, the trend of changes in design variables in each cluster can be easily grasped.
- the clustering unit 14 places the clustered model as shown in FIG. 36 on the space (plane) represented by the coordinate axes of the toe angle and camper angle. Then, clusters are sequentially sampled along a curve indicating a predetermined characteristic value change on the cluster group until the clustering unit 14 f. For example, clusters (10), (8), (15), (13), (1 along a curve (straight line) where the toe angle is approximately “-1” and the camper angle gradually increases. ), (2) are sampled in order. In other words, assuming a straight line in which only the camber angle changes, the cluster is sampled along this straight line. Subsequently, as shown in Fig.
- the principle extraction unit 15 can design each of the sampled clusters (10), (8), (15), (13), (1), (2). Draw the mean value of the variable on the drawing. If the average values of these design variables are plotted in order according to the sampling order described above, the design variables change as indicated by the wavy arrows. From this figure, the camber angle, which is the characteristic value, is greatly affected by changes in the X coordinate of the trailing arm joint P1, the Z coordinate of the lower arm joint P4, and the Z coordinate of the upper arm joint P3. It can be confirmed that it is affected.
- the X coordinate of the trailing arm joint P 1 and the Z coordinate of the lower arm joint P 4 are gradually decreased, and the Z coordinate of the upper arm joint P 3 is gradually increased.
- the toe angle is greatly influenced by the change of the Z coordinate of the trailing arm joint P 1.
- the camber angle is greatly affected by changes in the X coordinate of the trailing arm joint P1, the Z coordinate of the lower arm joint P4, and the Z coordinate of the upper arm joint P3.
- the sampling curve can be set in a so-called interactive manner while the operator looks at the display 10 8. Further, it can be automatically set by the analysis system according to the present embodiment. In other words, the analysis system detects the distribution of the cluster groups shown in Figures 36, 38, and 40, and automatically calculates characteristic curves such as envelopes of the cluster groups and straight lines that cross the cluster groups. In addition, clusters can be sampled along these characteristic curves. This enables automatic analysis.
- the present invention is not limited to the above-described configuration, and can be changed without departing from the spirit of the present invention.
- the present invention is not limited to the above design field.
- the present invention can be applied to a design field in which simulation calculation is possible.
- electronic circuit simulation, structural setup The present invention can be applied to a wide range of fields such as totals, software design, stock price prediction, and traffic congestion prediction.
- a software program for executing the above-described processing may be downloaded from a server or a download site and used as well as being installed in the computer.
- the program software may be installed from a storage medium such as a CD-ROM.
- an encrypted program may be distributed to users, and the decryption key may be notified only to users who have paid the price.
- any operating system for executing the program may be used, and any form of hardware for executing the program may be used.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Aviation & Aerospace Engineering (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP05799144.0A EP1811411A4 (en) | 2004-10-26 | 2005-10-25 | MULTI-VARIABLE MODEL LASER SYSTEM, PROCESS, AND PROGRAMMEDIUM |
US11/577,979 US7761267B2 (en) | 2004-10-26 | 2005-10-25 | Multi-variable model analysis system, method and program, and program medium |
JP2006542362A JP5002811B2 (ja) | 2004-10-26 | 2005-10-25 | 多変数モデル解析システム、方法、プログラム、およびプログラム媒体 |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2004310937 | 2004-10-26 | ||
JP2004-310937 | 2004-10-26 | ||
JP2005066086 | 2005-03-09 | ||
JP2005-066086 | 2005-03-09 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006046737A1 true WO2006046737A1 (ja) | 2006-05-04 |
Family
ID=36227969
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2005/019960 WO2006046737A1 (ja) | 2004-10-26 | 2005-10-25 | 多変数モデル解析システム、方法、プログラム、およびプログラム媒体 |
Country Status (4)
Country | Link |
---|---|
US (1) | US7761267B2 (ja) |
EP (1) | EP1811411A4 (ja) |
JP (1) | JP5002811B2 (ja) |
WO (1) | WO2006046737A1 (ja) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007087098A (ja) * | 2005-09-22 | 2007-04-05 | Nissan Motor Co Ltd | 最適化システム、最適化方法、最適化プログラム、及びプログラム媒体 |
JP2008165310A (ja) * | 2006-12-27 | 2008-07-17 | Yokohama National Univ | 多変数モデル解析システム、方法、プログラム、およびプログラム媒体 |
CN102046427A (zh) * | 2008-04-01 | 2011-05-04 | 克朗布股份有限公司 | 用来监视车辆驾驶过程的设备 |
US8600534B2 (en) | 2008-07-01 | 2013-12-03 | Airbus Operations Ltd | Method of designing a structure |
JP2014076701A (ja) * | 2012-10-09 | 2014-05-01 | Toyo Tire & Rubber Co Ltd | タイヤ設計方法、タイヤ設計用支援装置及びタイヤ設計用支援プログラム |
JP2014157491A (ja) * | 2013-02-15 | 2014-08-28 | Toyo Tire & Rubber Co Ltd | クラスタ分析方法、クラスタ分析装置及びコンピュータプログラム |
KR20190113924A (ko) * | 2017-03-27 | 2019-10-08 | 알리바바 그룹 홀딩 리미티드 | 채점 모델을 구축하고 사용자 신용을 평가하기 위한 방법 및 디바이스 |
JP2021005278A (ja) * | 2019-06-27 | 2021-01-14 | 富士通株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4815887B2 (ja) * | 2005-06-17 | 2011-11-16 | 日産自動車株式会社 | 情報処理装置及び情報処理用表示装置 |
AT9467U3 (de) * | 2007-06-14 | 2008-07-15 | Avl List Gmbh | Vorrichtung und verfahren zur simulation einer entwicklungsanlage |
AT10759U3 (de) * | 2009-04-23 | 2010-07-15 | Avl List Gmbh | Verfahren und vorrichtung zur verifizierung eines automatisierungssystems |
US9778636B2 (en) * | 2013-08-13 | 2017-10-03 | Siemens Product Lifecycle Management Software Inc. | Method for classifying and correlating simulation components to functional models |
US10346851B1 (en) * | 2016-07-05 | 2019-07-09 | Numerify, Inc. | Automated incident, problem, change correlation analysis system |
CN106250658B (zh) * | 2016-08-29 | 2019-07-19 | 国网冀北电力有限公司电力科学研究院 | 基于径向基函数的负荷开关电磁机构快速计算方法 |
US11163919B2 (en) * | 2018-12-26 | 2021-11-02 | Palo Alto Research Center Incorporated | Method and system for component-based design of physical systems |
CN111507003A (zh) * | 2020-04-20 | 2020-08-07 | 中国计量大学 | 基于质量相关动态特性提取的脱丁烷塔关键变量预测方法 |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000315218A (ja) | 1999-04-30 | 2000-11-14 | Honda Motor Co Ltd | 自動車の構造設計方法 |
AU4733601A (en) * | 2000-03-10 | 2001-09-24 | Cyrano Sciences Inc | Control for an industrial process using one or more multidimensional variables |
JP4836219B2 (ja) | 2001-05-31 | 2011-12-14 | 東洋ゴム工業株式会社 | コンピュータシミュレーションによるタイヤ設計方法及びタイヤ設計プログラム及びタイヤ設計装置 |
AU2002316267A1 (en) * | 2001-06-14 | 2003-01-02 | Rigel Pharmaceuticals, Inc. | Multidimensional biodata integration and relationship inference |
KR20040111456A (ko) * | 2002-04-04 | 2004-12-31 | 이시하라 산교 가부시끼가이샤 | 데이터 해석 장치 및 방법 |
US7596247B2 (en) * | 2003-11-14 | 2009-09-29 | Fujifilm Corporation | Method and apparatus for object recognition using probability models |
-
2005
- 2005-10-25 US US11/577,979 patent/US7761267B2/en active Active
- 2005-10-25 JP JP2006542362A patent/JP5002811B2/ja active Active
- 2005-10-25 WO PCT/JP2005/019960 patent/WO2006046737A1/ja active Application Filing
- 2005-10-25 EP EP05799144.0A patent/EP1811411A4/en not_active Ceased
Non-Patent Citations (4)
Title |
---|
"Program Product MIOS7/AS2 ETOILE/OP Tokei Bunseki Kino Kaisetsusho.", FIRST EDITION, HITACHI, LTD., 1993, pages 58 - 91, XP002999867 * |
QIANG GAN ET AL: "CAE o Mochiita Sharyo Seino Sekkei ni Okeru Sekkei Ryoiki o Shimesu Shinsekkei Shuho no Kaihatsu. (The development of the New Design Method Showing a Design Space in the Vehicles Performance Design using CAE)", THA JAPAN SOCIETY OF MECHANICAL ENGINEERS DAI 13 KAI SEKKEI KOGAKU SYSTEM BUMON KOENKAI KOEN RONBUNSHU., 30 October 2003 (2003-10-30), pages 289 - 292, XP002999866 * |
See also references of EP1811411A4 * |
UEDA TAICHIRO.: "Personal Computer o Mochiita Data Mining Katsuyo Kanzen Master Manual.", FIRST EDITION, NIPPON BIJINESU REPOTO KABUSHIKI KAISHA., 1998, pages 58 - 64, XP002999868 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007087098A (ja) * | 2005-09-22 | 2007-04-05 | Nissan Motor Co Ltd | 最適化システム、最適化方法、最適化プログラム、及びプログラム媒体 |
JP2008165310A (ja) * | 2006-12-27 | 2008-07-17 | Yokohama National Univ | 多変数モデル解析システム、方法、プログラム、およびプログラム媒体 |
CN102046427A (zh) * | 2008-04-01 | 2011-05-04 | 克朗布股份有限公司 | 用来监视车辆驾驶过程的设备 |
CN102046427B (zh) * | 2008-04-01 | 2013-10-16 | 克朗布股份有限公司 | 用来监视车辆驾驶过程的设备 |
US8600534B2 (en) | 2008-07-01 | 2013-12-03 | Airbus Operations Ltd | Method of designing a structure |
JP2014076701A (ja) * | 2012-10-09 | 2014-05-01 | Toyo Tire & Rubber Co Ltd | タイヤ設計方法、タイヤ設計用支援装置及びタイヤ設計用支援プログラム |
JP2014157491A (ja) * | 2013-02-15 | 2014-08-28 | Toyo Tire & Rubber Co Ltd | クラスタ分析方法、クラスタ分析装置及びコンピュータプログラム |
KR20190113924A (ko) * | 2017-03-27 | 2019-10-08 | 알리바바 그룹 홀딩 리미티드 | 채점 모델을 구축하고 사용자 신용을 평가하기 위한 방법 및 디바이스 |
KR102315497B1 (ko) * | 2017-03-27 | 2021-10-21 | 어드밴스드 뉴 테크놀로지스 씨오., 엘티디. | 채점 모델을 구축하고 사용자 신용을 평가하기 위한 방법 및 디바이스 |
US11188565B2 (en) | 2017-03-27 | 2021-11-30 | Advanced New Technologies Co., Ltd. | Method and device for constructing scoring model and evaluating user credit |
JP2021005278A (ja) * | 2019-06-27 | 2021-01-14 | 富士通株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
JP7288190B2 (ja) | 2019-06-27 | 2023-06-07 | 富士通株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
Also Published As
Publication number | Publication date |
---|---|
EP1811411A1 (en) | 2007-07-25 |
US20090132208A1 (en) | 2009-05-21 |
JP5002811B2 (ja) | 2012-08-15 |
JPWO2006046737A1 (ja) | 2008-05-22 |
EP1811411A4 (en) | 2014-05-07 |
US7761267B2 (en) | 2010-07-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2006046737A1 (ja) | 多変数モデル解析システム、方法、プログラム、およびプログラム媒体 | |
US20040153979A1 (en) | System and method for generating a two-dimensional yield map for a full layout | |
US6598211B2 (en) | Scaleable approach to extracting bridges from a hierarchically described VLSI layout | |
Mobley et al. | An Object Oriented Approach to Geometry Defeaturing for Finite Element Meshing. | |
Chen et al. | Triangulated manifold meshing method preserving molecular surface topology | |
JP2016224945A (ja) | 厚さ基準を用いたデータベースの問い合わせ | |
JP2018022476A (ja) | モフォロジー基準によるデータベースの照会 | |
US20070136044A1 (en) | Efficient simulation of dominantly linear circuits | |
Lotan et al. | Efficient maintenance and self-collision testing for kinematic chains | |
Barrera et al. | High-order mesh morphing for boundary and interface fitting to implicit geometries | |
US20170131183A1 (en) | Device, method, and program for crash simulation | |
EP1685509B1 (en) | System, method, and computer program product for determining wall thickness in a graphic model | |
JP4867314B2 (ja) | コンセプト設計支援装置,コンセプト設計支援プログラム及びコンセプト設計支援方法 | |
Anton | Voronoi diagrams of semi-algebraic sets | |
JP2008003819A (ja) | 交互作用検出装置,交互作用検出用プログラムが記録された媒体及び交互作用検出方法 | |
Zou et al. | Constraint-based beautification and dimensioning of 3D polyhedral models reconstructed from 2D sketches | |
JP4815887B2 (ja) | 情報処理装置及び情報処理用表示装置 | |
JP7029056B2 (ja) | 分割領域生成プログラム、分割領域生成装置、および分割領域生成方法 | |
Jiang et al. | Feature mining paradigms for scientific data | |
Chen et al. | Surface triangular mesh and volume tetrahedral mesh generations for biomolecular modeling | |
JP2008165310A (ja) | 多変数モデル解析システム、方法、プログラム、およびプログラム媒体 | |
Martorell et al. | High order unfitted finite element discretizations for explicit boundary representations | |
Shih et al. | Geometry and mesh generation for high fidelity computational simulations using non-uniform rational B-splines | |
JP5299471B2 (ja) | 情報処理用プログラム及び情報処理方法 | |
Csabai et al. | On the medial surface approximations of extrusions |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BW BY BZ CA CH CN CO CR CU CZ DK DM DZ EC EE EG ES FI GB GD GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV LY MD MG MK MN MW MX MZ NA NG NO NZ OM PG PH PL PT RO RU SC SD SG SK SL SM SY TJ TM TN TR TT TZ UG US UZ VC VN YU ZA ZM |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): GM KE LS MW MZ NA SD SZ TZ UG ZM ZW AM AZ BY KG MD RU TJ TM AT BE BG CH CY DE DK EE ES FI FR GB GR HU IE IS IT LU LV MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2006542362 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 11577979 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2005799144 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2005799144 Country of ref document: EP |