JP7454480B2 - Design support system and method - Google Patents

Description

The present invention relates to a design support system and method thereof.

In recent years, product design using CAD (Computer-Aided Design) has been frequently used. Guidelines for CAD design include, for example, rules from the perspective of ease of processing such as drilling and bending, ease of assembly such as welding and screw fastening, and ease of inspection and access to jigs. It includes many rules to be observed from the perspective of product maintenance, such as ease of use.

Therefore, it has been proposed that a computer equipped with CAD checks some or all of these rules and automatically determines whether a design error exists in the CAD model in light of guidelines. For example, Patent Document 1 discloses a system that executes a program including a combination of a plurality of functions in accordance with rules and notifies a designer if a CAD model does not satisfy a threshold.

JP2008-234011A

However, in order for a computer to check a CAD model based on rules, it is necessary to appropriately set parameters and threshold values for the function, but the system according to Patent Document 1 does not disclose this. I didn't even suggest it. Therefore, an object of the present invention is to provide a product design support system and method that can correctly check whether a CAD model is constructed in compliance with guidelines.

One of the representative design support systems of the present invention is a product design support system that causes a computer to check whether a CAD model of the product conforms to a guideline based on a determination rule. and a memory, the controller includes a geometric recognition function definition unit that sets a function for recognizing the form of the CAD model and generates a determination rule by executing a program stored in the memory; A parameter adjustment section that adjusts parameters based on the model check results and sets them in functions, and a rule check that checks the CAD model based on the function in which the adjusted parameters are set and identifies design violations. and an output section for outputting the check result to the notification device.

According to the present invention, a design support system that streamlines checking is created by constructing and evolving new judgment rules that can be applied to CAD models of various forms different from conventional ones, such as those that include newly adopted parts. Can be provided.

FIG. 2 is a functional block diagram for explaining a check function of the design support system according to the first embodiment of the present invention. 2 is a flowchart showing the procedure of a check operation in the design support system of FIG. 1. FIG. FIG. 2 is a functional block diagram showing the relationship between check functions in the design support system of FIG. 1 and hardware that implements them. 2 is a diagram visualizing and illustrating a CAD model to be checked in the design support system of FIG. 1. FIG. 2 is a flowchart showing the operation procedure of a parameter adjustment section and a rule check section in the design support system of FIG. 1. FIG. In the design support system of FIG. 1, this is an example of evaluation data stored in a parameter evaluation DB that is referred to during a check operation. FIG. 7 is a diagram visualizing and illustrating a CAD model to be checked in the design support system according to the second embodiment of the present invention. FIG. 2 is a functional block diagram for explaining a check function of a design support system according to a second embodiment of the present invention. 9 is a flowchart showing the operation procedure of a parameter adjustment section and a rule check section in the design support system of FIG. 8. 9 is a diagram illustrating a visualized correct answer data of a check result for the CAD model of FIG. 7 in the design support system of FIG. 8; FIG.

Hereinafter, a design support system according to an embodiment of the present invention will be described using the drawings. In Example 1, a case where the determination rule is quantified will be explained using FIGS. 1 to 7. In the second embodiment, a case where an unquantified determination rule is quantified will be explained using FIGS. 8 to 10.

In the first embodiment, a method for estimating input parameters for each function based on a quantified determination rule will be described. FIG. 1 is a functional block diagram for explaining a check function of a design support system according to a first embodiment of the present invention. Note that the determination rule is a rule that can be checked on a computer, and is a rule that can detect design mistakes from a CAD model being processed by a three-dimensional CAD system.

As shown in FIG. 1, the design support system 10 has a design guideline 1, a judgment rule definition section 2, a CAD model (CAD data) 3, a geometric shape recognition function database (hereinafter referred to as "geometric recognition function DB") as its checking function. ) 4, geometric recognition function module definition section (hereinafter referred to as "geometric recognition function definition section") 5, parameter evaluation DB 6, parameter adjustment section 7, rule check section 8, check result display section 9, and shape feature amount extraction section 16.

These functions are functions formed by a computer forming the design support system 10 executing a program, and a DB formed in a storage unit (memory) included in the computer. Note that the basic drawing functions in the design support system 10 are not shown in FIG. 1, and their explanation will be omitted.

The design guideline 1 includes design rules such as specifications of manufacturing equipment and tools, ease of assembly, processing limits, and standards such as laws and regulations and JIS. For example, the design guideline 1 includes the position regulations from the ends and bends of mechanical parts to holes, as well as the accessibility of manufacturing tools.

The judgment rule definition unit 2 is provided with an interface for defining judgment rules to be verified on the CAD model 3. This interface includes a user interface for the user to define design elements by referencing or importing the rules in Design Guideline 1, and a user interface for defining design elements by referring to or importing the rules in Design Guideline 1. Contains an interface for defining rules. The CAD model 3 is checked by the design support system 10 by applying a determination rule.

In the geometric recognition function DB 4, in addition to collecting a group of procedural functions (hereinafter referred to as "common functions") for recognizing geometric shapes included in the CAD model 3, a group of newly generated functions is also recorded. There is. Common functions can be called individually on a computer and are made into basic elements for recognizing geometric shapes, including functions for measuring distance. Note that a specific example of the common function will be described later. Further, the hardware of the computer constituting the design support system 10 will be described later using FIG. 3.

Design Guideline 1 includes a large number of rules to check. This design guideline 1 is defined in the determination rule definition section 2. In addition, by extracting and combining the common functions of the geometry recognition function DB 4 with respect to the design guideline 1 to be checked defined in the judgment rule definition unit 2, the geometry recognition function module definition unit (hereinafter referred to as “geometric recognition function definition unit ) 5, a checkable judgment rule is generated. The geometric recognition function definition unit 5 is a function for setting a function for recognizing the form of a CAD model and constructing a judgment rule for automatic checking by executing a program by a computer forming the design support system 10. .

The parameter evaluation DB 6 stores information indicating the correlation between the common function input to the geometric recognition function definition unit 5 and its suitable parameters, and the evaluation results obtained by checking. Here, the parameter to be set is, for example, a quantitative value represented by a search range or a reference dimension, but is not limited to this.

In addition, the check evaluation results include, for example, the setting values of the parameters for determining the geometric shape specified in the common function described above, the check time T and the violation part identification accuracy, which will be described later using FIG. be. Furthermore, information indicating the correlation between the check evaluation results, typified by the amount of noise N, the check time T, etc., and the parameter setting values is stored in the parameter evaluation DB 6.

The rule checking unit 8 inputs the common function and its suitable parameters, and checks the CAD model 3 to be checked. The common function used here is a component of the geometric recognition function defined in the geometric recognition function definition unit 5. As suitable parameters for the common function, the optimum parameters obtained by the parameter adjustment section 7 by searching the parameter evaluation DB 6 are used so as to improve the evaluation of the check result. Further, from the check results of the rule check section 8, evaluation results represented by the correct answer rate R and the check time T are obtained.

The check result display section 9 highlights the identified violation location and notifies the designer. Furthermore, the check result display section 9 updates the obtained evaluation results to the geometric recognition function DB 4 and parameter evaluation DB 6, thereby feeding them back to the next check. In other words, the design support system 10 updates the determination rules based on the evaluation of the check results, and continues to evolve appropriately so that the scope of application can be expanded.

If the shape feature represented by the dimensions and shape of the input CAD model 3 has a strong correlation with the parameter setting value, the shape feature extractor 16 extracts the feature from the CAD model 3, and the parameter adjuster 7 extracts the feature from the CAD model 3. It can also be used as input. The shape feature amount here includes, for example, the minimum (maximum) plate thickness included in the CAD model 3, the distribution of dimensions of parts, and feature information, but is not limited to these.

FIG. 2 is a flowchart showing the procedure of a check operation in the design support system 10 of FIG. Steps S1 to S8 shown in FIG. 2 are an example of a design support method, in which a computer equipped with a controller 22 (FIG. 3) and a memory (storage unit) applies judgment rules to the CAD models 3 and 30 to design a design. This is the operation procedure for extracting the violation points above. These steps S1 to S8 are check operation procedures that are mainly executed by the controller 22 of the computer that constitutes the design support system 10. Hereinafter, the operating entity without a subject is the computer controller 22, and more specifically, each functional unit formed therein.

At the start of the check, in a CAD model (data) reading step S1, the controller 22 reads the CAD model 3 to be checked into the storage section. In the determination rule definition step S2, determination rules to be checked are defined based on the design guideline 1. In other words, whether or not each design element is a violation is defined as a determination rule. In the common function storage step S3, the controller 22 stores the common function used for verification of each defined design element in the geometric recognition function DB (storage unit) 4.

In the function definition step S4, the judgment rule definition unit 2 of the controller 22 queries the geometric recognition function DB4, and selects the optimum one for identifying the violation location of the CAD model 3 from among the judgment rules defined in the judgment rule definition step S2. Find and define common functions. In this way, in step S4, the optimal common function defined in step S2 and stored in step S3 is searched from the geometric recognition function DB4, and the obtained optimal common functions are combined to determine the optimal common function based on the determination rule. Build a geometry recognition function for automatic checking.

Here, specific examples of common functions include geometric feature search functions such as searching for holes and edges, shape feature calculation functions such as measuring the distance between surfaces, and shape generation functions such as surface creation and point creation. , numerical calculation functions for calculating four arithmetic operations, maximum values, minimum values, etc., but are not limited thereto.

In the evaluation result correlation storage step S5, the controller 22 stores the correlation between the parameters input to the common function and the evaluation results obtained by evaluating the check results including at least the correct answer rate R or the calculation time T in the parameter evaluation DB 6. . In step S5, the parameter evaluation DB 6 stores, for example, the correlation between the numerical value of the parameter corresponding to the input of the common function defined by the function and the evaluation result such as the correct answer rate R of the check result and the check time T. There is.

In the parameter adjustment step S6, the parameter adjustment unit 7 queries the parameter evaluation DB 6 stored in the parameter evaluation DB 6, and inputs and adjusts the common function searched based on the evaluation result and its suitable parameters. That is, in step S6, the parameter adjustment unit 7 adjusts the parameters input to the common function to appropriate values. Although there may be one type of parameter processed in step S6, there are usually multiple types.

In the check result evaluation step S7, the parameters adjusted by the parameter adjustment unit 7 in step S6 are input to the common function defined by the geometric function definition unit 5 in step S4, and the determination rule generated is applied to the CAD models 3, 30. , evaluate the check results. In this step S7, CAD is performed based on the determination rule constructed by inputting the function stored in the geometric recognition function DB 4 in step S3 and the parameter value stored in the parameter evaluation DB 6 in step S5 to the parameter adjustment section 7. Check model 3. In step S7, the checked results are evaluated. Evaluation indicators for check evaluation include, but are not limited to, the above-mentioned accuracy rate R and calculation time T.

If the evaluation result in step S7 does not satisfy the desired accuracy rate R or calculation time T (No), the parameter values are adjusted again in step S6. These steps are repeated, and when the evaluation of the check results satisfies the correct answer rate R and the calculation time T (Yes), the process proceeds to check result display step S8, the check results are displayed, and the check ends. Through the above flow, a checking method is established in which leakage F and noise N are reduced from the check results, and calculation time T is also shortened.

FIG. 3 is a functional block diagram showing the relationship between the check functions in the design support system 10 of FIG. 1 and the hardware that implements them. In FIG. 3, the design support system 10 includes a controller 22, which is a computer including an input section 20 and a display section 21. The input unit 20 and the display unit 21 provide the user with an operating environment using a GUI ( graphical user interface).

A control unit 23 , an input interface (hereinafter referred to as input I/F) 24 , a display control unit 25 , a main storage device 26 , and an auxiliary storage device 27 are connected to the controller 22 via a data bus 28 . The control unit 23 is the central part of the controller 22 in the design support system 10, and controls all of the drawing function and checking function (not shown). In particular, the control unit 23 as a checking function operates as if the geometric recognition function definition unit 5 causes the parameter adjustment unit 7 to generate a rule that can be checked, and causes the rule checking unit 8 to automatically check the rule. However, the division of functions between the control unit 23 and the geometric recognition function definition unit 5 is for convenience of explanation, and is not limited to this. The input I/F 24 takes in the CAD model and determination rules input from the input unit 20. The display control section 25 controls the display section 21 .

The main storage device 26 stores a CAD model 3, a design guideline 1, a judgment rule definition section 2, a parameter adjustment section 7, a rule check section 8, and information and programs for making these function. The auxiliary storage device 27 stores information and programs of the geometric recognition function DB4 and parameter evaluation DB6.

Note that part or all of the computer hardware of the controller 22 may be replaced by a DSP (Digital Signal Processor), an FPGA (Field-Programmable Gate Array), a GPU (Graphics Processing Unit), or the like. Further, part or all of the hardware may be centralized or distributed on servers on a network, placed in the cloud, and shared by multiple users via the network.

FIG. 4 is a diagram visualizing and illustrating the CAD model 3 to be checked in the design support system 10 of FIG. 1. As shown in FIG. The CAD model 3 shown in FIG. 4 is a part of a product that is fixed to another component with bolts 33 via a fixture 32 for fixing a pipe 31. The design support system 10 automatically checks whether the bolt insertion hole has an appropriate size for the bolt 33 among these.

FIG. 5 is a flowchart showing the operation procedure of the parameter adjustment section 7 and rule check section 8 in the design support system 10 of FIG. More specifically, FIG. 5 shows an operation procedure in which the design support system 10 automatically checks bolt holes in the CAD model 3 shown in FIG. 4. In this embodiment, the geometric recognition function definition section 5 is uniquely determined, and the operation flow of the parameter adjustment section 7 and the rule check section 8 is shown together, but the modules constituting the geometry recognition function definition section 5 can be changed as appropriate. and is not limited to this.

First, when a check is started, initial values of parameters are set in step S11. Thereafter, a bolt is recognized on the CAD model 3 in step S12. As the bolt recognition method in step S12, bolts may be identified based on their shape, but it is preferable that the shape feature extracting unit 16 utilizes part names and attribute information to recognize them.

That is, since it is possible to automatically extract how many bolts of which size are included in the CAD model 3, it is desirable to utilize such information. In step S13, the central axis of the bolt is generated. In step S14, holes included on the CAD model 3 are searched, and in step S15, holes near the bolts are identified.

In step S16, the hole diameter of the hole obtained in step S15 is measured, and in step S17, the hole diameter is determined. Here, if there is no violation (No), the check ends. If there is a violation (Yes), it is evaluated whether the result detected as a violation in step S18 is correct, and the process proceeds to correct answer rate evaluation step S19. In step S19, if the evaluation result represented by the accuracy rate R of the detection result and the check time T satisfies the predetermined performance (Yes), the check ends.

On the other hand, if the evaluation result in step S19 does not satisfy the predetermined performance (No), the process returns to parameter setting step S11. In step S11, the parameter evaluation DB 6 (FIGS. 1 and 3) is consulted and the parameters are reset. This process is repeated, and if the accuracy rate R or check time T for the check satisfies the target performance (Yes), the check ends and the violations are highlighted (not shown).

FIG. 6 is an example of evaluation data stored in the parameter evaluation DB 6 that is referred to during a check operation in the design support system 10 of FIG. 1. FIG. 6 is a graph showing the evaluation results obtained by executing the hole search function, that is, the hole recognition step S14 in FIG. 5, for the CAD model 3 in FIG. Further, FIG. 6 also shows hole detection results 43. The horizontal axis in FIG. 6 indicates the ratio of hole search reference hole diameter/maximum bolt diameter from 0 to 10. The left vertical axis in FIG. 6 is the bolt hole correct answer rate R (%), which indicates the correct answer rate R (%) at which bolt insertion holes were detected from the check results. The right vertical axis in FIG. 6 is the dimensionless calculation time T, which indicates the time required for the check. A characteristic 41 in FIG. 6 represents the relationship between the reference hole diameter and the bolt hole accuracy rate R in the hole search function. A characteristic 42 in FIG. 6 represents the relationship between the reference hole diameter and the bolt hole correct answer rate R time T in the hole search function.

FIG. 6 will be explained in more detail. The ratio shown on the horizontal axis of FIG. 6 is the ratio of the hole search reference hole diameter, which is an input parameter set in the hole recognition step S14, and the maximum bolt diameter, and if the ratio is twice the maximum bolt diameter M10, the hole search is performed. If the reference hole is 20 mm and the ratio is 10 times, the hole search reference hole is 100 mm. The technical significance of "Hole search standard hole diameter/Maximum bolt diameter" shown on the horizontal axis of Figure 6 is explained later in the [Supplementary] column, "The hole through which the bolt passes must be within a range of 〇mm to △mm larger than the bolt diameter. Based on design guidelines such as Judgment rules based on the design guidelines to be checked are defined for each design element to determine whether or not it violates the design guidelines. One of the parameters input to the common function forming this determination rule is shown on the horizontal axis of FIG.

FIG. 6 shows that the evaluation of the check result based on the determination rule generated by the common function changes depending on the parameters input to the common function. That is, FIG. 6 shows the correlation between arbitrarily input parameters and the evaluation of the check results. The correlation of this evaluation result is stored in the parameter evaluation DB 6 as a parameter evaluation, and is provided for reference by the parameter adjustment section 7. The parameter adjustment unit 7 adjusts the parameters to obtain a high evaluation based on the evaluation of the referenced check results. In FIG. 6, the higher the bolt hole accuracy rate R and the smaller the dimensionless calculation time T, check omission F, and noise N, the higher the evaluation.

The bolt hole accuracy rate R shown on the left vertical axis in Figure 6 means that when the number of holes detected is larger than the number of correct answers, there is a lot of noise N. rate) becomes the correct answer rate R. On the other hand, if the number of holes detected is smaller than the number of correct answers, it means that a leakage F has occurred, and the value (recall rate) obtained by dividing the number of holes detected by the number of correct answers is the correct answer rate R. Here, evaluation indicators such as a recall rate representing the leakage F and a precision rate representing the amount of noise N may be applied separately, and the evaluation indices for the check results are not limited to these.

In the hole search function shown in the hole recognition step S14 and the bolt vicinity hole search step S15, when the reference hole diameter is increased, the range of holes to be acquired also increases, and the calculation time T becomes longer. In addition, pipes with large hole diameters are also acquired, and the noise N due to non-bolt holes increases. On the other hand, if the reference hole diameter is made smaller, although the calculation time T becomes shorter, bolt hole leakage F occurs, which causes check leakage F.

In the above-described hole search function (S14, S15), it can be seen that the current optimum value is approximately twice the maximum bolt diameter Dmax included in the CAD model 3 to be checked. In the example of FIG. 6, a method of searching for an optimal value for one parameter is shown, but there may be a plurality of types of parameters, or an input parameter that is commonly input to a plurality of common functions.

In Example 2, regarding a design rule for which the judgment rule is not quantified, a method is explained in which the input parameters and threshold values of a common function are determined and the judgment rule is constructed using data for which violations are known from past design data. do.

FIG. 7 is a diagram visualizing and illustrating a CAD model (CAD data) 30 to be checked in the design support system 20 according to the second embodiment of the present invention. That is, FIG. 7 shows a part of the CAD model 30 targeted in the second embodiment. Here, a bolt 33 passes through a bolt-penetrating component 14 having a bend, and there is a design guideline 1 that "checks whether a space for inserting a tool for tightening the bolt is secured."

When constructing judgment rules that can be checked on the CAD model 30 from the design guideline 1, it is difficult for a computer to automatically and accurately reproduce and evaluate the shape and operating range of the tool. Therefore, in the design support system 20, a method is adopted in which a shape is created that simulates the insertion space for the fastening tool at the top of the hexagonal part of the bolt 33, and if other parts do not interfere with this space, the evaluation is passed. , its shape is simulated by a cylinder.

Here, if the dimensions of the evaluation cylinder are set large, a large number of interfering parts will be detected with respect to the evaluation cylinder, and the noise N will increase. On the other hand, if the dimensions of the evaluation cylinder are set small, interference between the evaluation cylinder and other parts cannot be identified, and a check failure F occurs. Therefore, in the design support system 20, correct data is created based on past manufacturing data of positions where bolts could be fastened and positions where bolts could not be fastened, and the check results are compared with the correct data, and the fastening tool Adjust the insertion space, that is, the dimensions of the evaluation cylinder.

Figure 8 is a functional block diagram for explaining the check function of the design support system 20 according to the second embodiment of the present invention. As shown in Figure 8, among the check functions of the design support system 20, the design guideline 1, the judgment rule definition unit 2, the geometric recognition function DB 4, the geometric recognition function definition unit 5, the parameter evaluation DB 6, the parameter adjustment unit 7, the rule check unit 8, and the check result display unit 9 are common to the design support system 10 according to the first embodiment shown in Figure 1.

Here, the CAD model (CAD data) 30 to be checked is the one described above using FIG. In the design support system 20, the correct data on the presence or absence of violations created based on the past manufacturing data described above from the CAD model 30, that is, the correct data to be checked 17, is input to the parameter adjustment unit 7. Here, the parameter adjustment unit 7 uses the check target correct answer data 17 to adjust the parameters quickly and accurately.

FIG. 9 is a flowchart showing the operation procedure of the parameter adjustment section 7 and rule check section 8 in the design support system 20 of FIG. FIG. 9 shows an operating procedure for checking for harmful interference with the bolt fastening tool insertion space. That is, the parameter setting in step S11 and the bolt recognition in step S12 in the second embodiment shown in FIG. 9 are the same as in the first embodiment shown in FIG. 5. Next, in step S21 of FIG. 9, an evaluation cylinder is created above the hexagonal head using bolts. Although the shape of the tool is simulated here as a cylinder, it is not limited to this shape, and may be a rectangular parallelepiped, a cone, or a semicircle.

Next, in step S22, parts near the bolt are searched for. Next, in step S23, it is checked whether there is any interference between parts near the bolt and the cylindrical surface for evaluation. Next, if it is determined in step S24 that there is interference (Yes), the process proceeds to step S25, and if it is determined that there is no interference (No), the check ends. In step S25, it is evaluated whether the check result is correct as a violation location by collating the correct answer data.

Next, in step S26, if there is a discrepancy between the check result and the correct data (No), the process advances to step S27. In step S27, the parameter adjustment unit 7 queries the parameter evaluation DB 6 and adjusts the parameters, and then returns to step S11 again to reset the parameters. This series of steps is repeated, and when the deviation between the check result and the correct data becomes the minimum, it is determined that the process is OK (Yes) in step S26, and the value of the parameter is determined. As a result, the determined parameter values are input into the common function, and the determination rule configured by the common function is applied, and the check on the CAD model 30 is completed.

FIG. 10 is a diagram visualizing and illustrating the correct data of the check results for the CAD model 30 in FIG. 7 in the design support system 20 in FIG. 8. FIG. 10 shows a constraint condition (bad sample) 12 under which a tool cannot be inserted, and an evaluation cylinder 13 (good sample) displaying conditions for obtaining a good judgment. Here, a design state related to a bolt in which a bolt tool has not been inserted is set as a constraint condition 12 in the correct data to be checked 17 (FIG. 8).

It is assumed that the dimensions of the evaluation cylindrical surface 11 can be determined in the parameter adjustment section 7 so that the constraint condition 12 in which the tool cannot be inserted and the evaluation cylinder 13 can be detected. Here, the greater the number of correct data, the higher the parameter estimation accuracy, and usually about several hundred pieces are required, but the number is not limited to this. As described above, in the design support system 20, the parameter adjustment unit 7 can automatically check the rules even for the unquantified design guideline 1 by utilizing past performance data.

[supplement]
In CAD design, it is desirable to comply with design guidelines. For example, design guidelines related to ease of assembly include ease of processing such as drilling and bending, and ease of assembly such as welding and screw fastening, which are problems during product manufacturing. In addition, design guidelines are also exemplified that specify issues such as ease of inspection, ease of procuring jigs used therein, and ease of compatibility with jigs, which are issues during product maintenance.

For example, in relation to the "ratio of hole search standard hole diameter/maximum bolt diameter" shown on the horizontal axis of Figure 6, "the hole through which the bolt is passed should be in the range of 〇mm to △mm larger than the bolt diameter". Design guidelines exist. If you apply the design guidelines mentioned above, such as ``Make the hole through which the bolt goes in a range from 〇mm to △mm larger than the bolt diameter,'' you can combine functions such as searching for bolts and holes, and measuring the hole diameter. Automatic checking becomes possible by building rules.

However, even with the same guideline, the size and structure differ depending on the product type, so it is necessary to change the search range and shape dimensions depending on the product type. For example, in the case of a hole search function, if a reference hole diameter is set, a hole smaller than the reference hole diameter is searched for. Here, if the reference hole diameter is set smaller than the bolt diameter included in the CAD model, the bolt insertion hole will not be checked, and the CAD model will not play a role in discovering design errors.

On the other hand, if the reference value of the diameter of the hole to be searched around the bolt is set to a large value, the calculation time required for the search will increase, and large holes other than the bolt insertion hole will be detected, which will generate noise. Noise refers to output that does not correspond to the desired answer. In this way, each function for creating judgment rules requires frequent trial and error to set input parameter values (also simply referred to as "parameters"), such as ranges and dimensions, which is time consuming. It required

On the other hand, there are rules in the design guidelines that are not quantified, such as ``securing a space for the insertion of a tool to tighten bolts.'' In that case, it is necessary to estimate not only the input parameters for each function of the check process but also the threshold values used for determination.

The design support system (this system) 10, 20 according to the embodiment of the present invention can be summarized as follows.
[1] This system 10, 20 allows a computer equipped with a controller 22 and a menu to execute a program stored in a memory (storage unit) to ensure that the CAD model 3, 30 of the product conforms to the guidelines. This is a product design support system that uses a computer to check based on judgment rules. The controller 22 forms a geometric recognition function definition section 5, a parameter adjustment section 7, a rule check section 8, and an output section.

The geometric recognition function definition unit 5 sets a function for recognizing the form of a CAD model and generates a determination rule. The parameter adjustment unit 7 adjusts parameters based on the check results of the CAD model and sets them in the function. That is, the parameter adjustment unit 7 inputs arbitrary parameters into the common function to extract new design violations based on various types of CAD models different from conventional ones, such as those that include newly adopted parts. The judgment rule is applied to the CAD models 3, 30, and the accuracy of the check result is adjusted based on the evaluation of the check result.

The CAD model is checked based on the function in which the adjusted parameters are set, and design violations are identified. That is, the rule check unit 8 inputs the parameters adjusted by the parameter adjustment unit 7 into a common function so as to be compatible with various types of CAD models that are different from conventional ones, such as those that include newly adopted parts. Performs a rule check. The output unit causes the check result display unit 9 to display the check result of the rule check performed by the rule check unit 8. The systems 10 and 20 configured in this manner can construct and evolve new judgment rules that can be applied to various types of CAD models that are different from conventional ones, such as those that include newly adopted parts.

[2] In the present systems 10 and 20, it is preferable that the controller 22 further includes a determination rule definition section 2, a geometric recognition function DB4, and a parameter evaluation DB6. The determination rule definition unit 2 defines, as a determination rule, whether or not each design element violates the design guideline 1 to be checked. The geometric recognition function DB 4 stores common functions that constitute determination rules for each design element based on the design guideline 1.

In addition, the geometric recognition function definition unit 5 is capable of specifying violation locations in the CAD models 3 and 30 by a combination of common functions obtained by querying the geometric recognition function DB 4 for determination rules. Define the common functions that make up the judgment rules.

Further, the parameter evaluation DB 6 stores correlations of parameters with respect to evaluations of check results based on determination rules generated using a common function input with arbitrary parameters. That is, the parameter evaluation DB 6 stores correlations of parameters with check results based on judgment rules generated by inputting arbitrary parameters into a common function for various types of CAD models different from conventional ones, such as those including newly adopted parts. do. That is, the parameter evaluation DB 6 stores the correlation between the parameters input to the common function and the evaluation results obtained by evaluating the check results including at least the correct answer rate R or the calculation time T according to the parameters.

The present systems 10 and 20 solve the following first and second problems by newly constructing and evolving determination rules.
1) As a first problem, it is desirable to shorten the trial-and-error time required to determine parameters that constitute a determination rule.
2) As a second issue, in order to construct and evolve decision rules for automatic checking, it is necessary to estimate parameters and threshold values that quantify the decision rules and adjust them with higher precision.

In order to solve the first and second problems, the systems 10 and 20 include a determination rule definition section 2, a geometry recognition function DB 4, a geometry recognition function definition section 5, a parameter evaluation DB 6, and a parameter adjustment section 7. , a rule check section 8, and a check result display section 9.

The determination rule definition unit 2 defines whether or not each design element is a violation as a determination rule. The geometric recognition function DB4 stores common functions that constitute determination rules. The geometric recognition function definition unit 5 defines a violation part of the CAD model 3 to be specified by a combination of common functions obtained by querying the geometric recognition function DB 4 for determination rules. The parameter evaluation DB 6 stores the correlation between the parameters input to the common function and the evaluation results obtained by evaluating the check results including at least the correct answer rate R or the calculation time T depending on the parameters.

The parameter adjustment section 7 queries the parameter evaluation DB section (storage section) 6, and inputs and adjusts the common function searched based on the evaluation result and its suitable parameters. Thereby, the present systems 10 and 20 can solve the first problem of requiring time for trial and error to determine the parameters that constitute the determination rule. Furthermore, in order to construct and evolve judgment rules for automatic checking, the systems 10 and 20 need to estimate parameters and thresholds that quantify the judgment rules and adjust them with higher precision. can also solve problems.

The rule check unit 8 performs a rule check using a determination rule generated by inputting the parameters adjusted by the parameter adjustment unit 7 into the function defined by the geometric recognition function definition unit 5. The check result display section 9 causes the display section 21 to display the check results checked by the rule check section 8.

According to the systems 10 and 20, the parameters that are input to the geometric recognition function in the parameter adjustment section 7 can be optimized for dimensions and threshold values that differ from product to product. As a result, the present systems 10 and 20 have the effect of reducing noise N and check omissions F from the check results and shortening the calculation time T for dimensions and threshold values that differ from product to product.

[3] The present system 10 may further include a shape feature extraction unit 16. The shape feature extraction section 16 inputs the shape feature included in the CAD model 3 to the parameter adjustment section 7, thereby making the parameters more appropriate. Thereby, the system 10 can further enhance the effect of [2] above.

[4] In the present system 10 described in [3] above, the shape feature extraction unit 16 extracts information representing the dimension information and distribution of the parts included in the CAD model 3 as shape features, and extracts the extracted shape features. It is preferable to input this into the parameter adjustment section 7. Thereby, the present system 10 can optimize the parameters in accordance with various types of CAD models that are different from conventional ones, such as those that include newly adopted parts. As a result, the present system 10 can further enhance the effect of [3] above.

[5] In the present system 10 described in [3] above, it is preferable that feature information included in the CAD model 3 is also extracted and specified as the shape feature amount. As a result, according to the present system 10, parameters can be optimized more precisely using the feature information extracted from the CAD model 3, so that the effect of [3] above can be further enhanced.

[6] In the present system 10 described in [3] above, it is preferable that attribute information of the part is also specified in the shape feature amount. As a result, the present system 10 can more quickly optimize parameters using the attribute information extracted from the CAD model 3, so that the effect of [3] above can be further enhanced.

[7] In the systems 10 and 20 according to the second embodiment shown in FIG. 8, the parameter evaluation DB 6 stores the accuracy rate R, which is the ratio of noise N or check omission F included in the check results as shown in FIG. , and information indicating the correlation between the values of the parameters and the values of the parameters is stored.

The parameter adjustment unit 7 searches the parameter evaluation DB 6 and can search for and obtain optimal parameters that have been highly evaluated as shown in FIG. As a result, the present systems 10 and 20 can further enhance the effect of [2] above.

[8] In the systems 10 and 20 described in [2] above, the parameter evaluation DB 6 stores the evaluation results of the check time T required for rule checking and the parameter values as shown in FIG. Information indicating the correlation between and is stored. That is, based on empirical rules or the like, parameter values that can advantageously shorten the calculation time T are stored in advance in the parameter evaluation DB 6. According to the present systems 10 and 20, the parameter adjustment unit 7 reads out and applies the optimum value of the parameter from the parameter evaluation DB 6, thereby reducing the check time T, that is, the calculation time T, among the effects of [2] above. It is possible to further shorten the process and speed up the determination of the parameters that constitute the judgment rule.

DESCRIPTION OF SYMBOLS 1... Design guideline, 2... Judgment rule definition part, 3, 30... CAD model (CAD data), 4... Geometric recognition function DB, 5... Geometric recognition function definition part, 6... Parameter evaluation DB, 7... Parameter adjustment part 7 , 8... Rule check section, 9... Check result display section, 10, 20... Design support system, 11... Cylinder for evaluating bolt fastening tool insertion space, 12... Example of bolt position that violates correct answer data, 13... Correct answer Example of a bolt position that does not violate the data, 14... Part that the bolt penetrates, 16... Shape feature extraction section, 17... Correct data to be checked (of the violation location), 20... Input section, 21... Display Part, 22...Controller, 23...Control unit, 24...Input interface, 25...Display control unit, 26...Main storage device, 27...Auxiliary storage device, 28...Data bus, 31...Piping, 32...Fixing jig, 33 ...Bolt, 41...Characteristics representing the relationship between the reference hole diameter and bolt hole accuracy rate R in the hole search function, 42...Characteristics representing the relationship between the reference hole diameter and bolt hole accuracy rate R time T in the hole search function, 43... Hole search results in the CAD model to be checked, Dmax... Maximum bolt diameter, R... Bolt hole correct answer rate, T... Check time (dimensionalless calculation time)

Claims (13)

The product design support system causes a computer to check whether a CAD model of the product conforms to a guideline based on a determination rule,
The computer includes a controller, a memory,
The controller executes a program stored in the memory, thereby
a geometric recognition function definition unit that sets a function for recognizing the form of the CAD model and generates the determination rule;
a parameter adjustment unit that adjusts parameters based on the check results of the CAD model and sets them in the function;
a rule checking unit that checks the CAD model based on the function to which the adjusted parameters are set, and identifies design violations;
an output unit that outputs the check result to a notification device;
a shape feature amount extraction unit that inputs shape feature amounts included in the CAD model into the parameter adjustment unit to optimize the parameters;
form,
Design support system. The shape feature amount is information representing dimension information and distribution of parts included in the CAD model,
The design support system according to claim 1 . The shape feature amount also specifies feature information included in the CAD model.
The design support system according to claim 1 . The shape feature amount also specifies attribute information of the part.
The design support system according to claim 1 . The product design support system causes a computer to check whether a CAD model of the product conforms to a guideline based on a determination rule,
The computer includes a controller, a memory,
The controller executes a program stored in the memory, thereby
a geometric recognition function definition unit that sets a function for recognizing the form of the CAD model and generates the determination rule;
a parameter adjustment unit that adjusts parameters based on the check results of the CAD model and sets them in the function;
a rule checking unit that checks the CAD model based on the function to which the adjusted parameters are set, and identifies design violations;
an output unit that outputs the check result to a notification device;
form,
The controller includes:
a determination rule definition unit that defines whether or not each design element is a violation as the determination rule;
a geometric recognition function DB in which common functions constituting the determination rule are stored for each of the design elements based on design guidelines to be checked;
a parameter evaluation DB that stores correlations of the parameters with respect to evaluations of the check results according to the judgment rules generated by the common function into which arbitrary parameters are input;
Furthermore,
The geometric recognition function definition unit is configured to define the common function that configures the determination rule so that the violation location of the CAD model can be identified by a combination of the common functions obtained by querying the determination rule with the geometric recognition function. define a function,
The parameter evaluation DB stores a correlation between the parameter input to the common function and an evaluation result obtained by evaluating the check result, which includes at least a correct answer rate or calculation time according to the parameter.
The parameter adjustment unit adjusts the correlation so as to improve the accuracy of the check result based on the evaluation of the check result using the common function and the parameter extracted by searching the parameter evaluation DB in order to identify the violation location. Adjust based on gender,
Design support system. The parameter evaluation DB stores information indicating a correlation between the correct answer rate, which is the ratio of noise or omissions included in the check result, and the value of the parameter.
The design support system according to claim 5 . The parameter evaluation DB stores information indicating a correlation between the evaluation result of the check time required for the check and the value of the parameter;
The design support system according to claim 5 . A design support method in which a computer equipped with a controller and a memory applies judgment rules to a CAD model to extract design violations, the method comprising:
By the computer executing the program stored in the memory,
The controller includes:
Define a common function that constitutes the judgment rule,
Applying the judgment rule generated by inputting an arbitrary parameter to the common function to the CAD model and adjusting the parameter based on the evaluation of the check result of the rule check,
inputting the adjusted parameters into the common function and displaying the check results based on the determined determination rule ;
optimizing the parameters using shape feature amounts included in the CAD model;
Design support method. The shape feature amount is information representing the dimensional information and distribution of CAD models of various forms different from conventional ones, such as including newly adopted parts.
The design support method according to claim 8 . The shape feature amount is feature information or part attribute information,
The design support method according to claim 8 . A design support method in which a computer equipped with a controller and a memory applies judgment rules to a CAD model to extract design violations, the method comprising:
By the computer executing the program stored in the memory,
The controller includes:
Define a common function that constitutes the judgment rule,
Applying the judgment rule generated by inputting an arbitrary parameter to the common function to the CAD model and adjusting the parameter based on the evaluation of the check result of the rule check,
inputting the adjusted parameters into the common function and displaying the check results based on the determined determination rule;
The controller includes:
a CAD model reading step of reading the CAD model;
a determination rule definition step of defining whether or not each design element is in violation based on a design guideline to be checked as the determination rule;
a common function storing step of storing a common function used for verification of each of the defined design elements in the memory;
a function definition step of querying the memory and searching and defining the common function that is most suitable for identifying the violation location of the CAD model from among the judgment rules defined in the judgment rule definition step;
an evaluation result correlation storage step for storing the correlation between the parameters input to the common function and the evaluation results obtained by evaluating the check results including at least the correct answer rate or the calculation time;
a parameter adjustment step of searching the memory to identify the violation location and adjusting based on the correlation so as to improve the accuracy of the check result using the extracted common function and the parameter;
Check result evaluation of evaluating the check result obtained by applying the judgment rule generated by inputting the parameter adjusted in the parameter adjustment step to the common function defined in the function definition step to the CAD model. step and
a check result display step for displaying the check results evaluated in the check result evaluation step;
has,
Design support method. In the parameter adjustment step, optimal information is read from the memory out of the information indicating the correlation between the correct answer rate, which is the ratio of noise or check omission included in the check result, and the parameter input to the common function. to use,
The design support method according to claim 11 . In the parameter adjustment step, among the information indicating the correlation between the evaluation result of the check time and the parameter input to the common function, information with an optimal evaluation result is read from the memory and used.
The design support method according to claim 11 .

Images