JP5777491B2 - Deformation prediction formula setting method for resin bumper and deformation prediction method for resin bumper - Google Patents

Description

  The present invention relates to a method for setting a deformation prediction formula for a bumper generated when a resin bumper is attached to a vehicle body, and a deformation prediction method for a resin bumper.

  In recent years, attempts have been made to make exterior parts of automobiles such as bumpers made of resin for the purpose of improving fuel efficiency associated with weight reduction, and many of them have been mounted. However, when a relatively large part such as a bumper is made of a resin, when the bumper is attached to the vehicle body, the bumper is often deformed so as to hang down from the vehicle body due to its own weight. For this reason, for example, a gap between a peripheral part such as a hood, a fender, and a headlamp and the bumper is wider than expected, and there is a risk of causing an assembly failure such as a deterioration in design surface quality.

  In order to solve the above problems, the final design and creation of the product was conventionally performed by repeating design changes and mold corrections. However, this method is very time-consuming and expensive, so prototypes and corrections are required. The construction of a new design method to replace this process is desired.

  Here, for example, in Patent Document 1 below, not only the deformation amount at the time of molding a resin part such as a bumper, but also the deformation amount (sagging amount) generated when assembled to the support is taken into account the deformation amount due to its own weight. A method for predicting by CAE analysis is disclosed.

JP 2007-38600 A

  As described above, although a prediction technique using CAE analysis has been proposed, the prediction accuracy is still insufficient, and there is a problem that reliability and certainty are lacking. Had. In other words, there is often a large gap between the amount of deformation that occurs in the actual product and the analysis result (predicted amount), and as a result, it is necessary to repeat trial manufacture and its (mold) correction. It was happening. In addition, the above analysis mainly requires a great deal of time for creating and modifying an analysis model, and the time required for prediction is also long.

  In view of the above circumstances, a technical technique that should be solved by the present invention is to build a technology for predicting the deformation amount of a bumper mounted on a vehicle body with excellent reliability and certainty, thereby reducing the time required for final design. Let it be an issue.

  The solution of the technical problem is achieved by the method for setting the deformation prediction formula of the resin bumper according to the present invention. That is, this setting method is a method of setting a prediction formula for predicting the amount of deformation of the bumper that occurs when the resin bumper is attached to the vehicle body, and is a drawing element of the bumper in a plurality of vehicle models designed in the past. A correlation obtaining step for obtaining a correlation between the data and the amount of deformation at the time of attachment, and a regression analysis step for performing regression analysis on a plurality of higher-order drawing elements showing relatively high correlations. When setting the prediction formula for the amount of deformation during installation, the drawing element data is stratified according to whether the bumper is fastened to the radiator support in the longitudinal direction of the vehicle body or when it is fastened in the vertical direction. The correlation with the deformation amount at the time of installation is obtained for each element data, and the sign of the correlation showing a relatively high value is adjusted in view of the technical relationship between the corresponding drawing element and the deformation amount at the time of installation. To whether to determine which is only characterized with a point of performing a regression analysis on the determined drawing elements to be aligned.

  The present invention is characterized by setting a prediction formula for predicting the deformation amount of the resin bumper when the vehicle body is mounted using a so-called SQC method. In other words, the statistical quality control method here refers to the results (quantitative results) that can occur in a newly produced vehicle type from the results (data) actually measured in the past vehicle type and a plurality of drawing elements of the vehicle type. ) And is applied to predict the amount of deformation when the bumper is installed. Here, when the present inventors applied the above-mentioned quality control method to the prediction of the deformation amount at the time of mounting the resin bumper, if the conventional quality control method is applied as it is, the deformation amount at the time of mounting the bumper is still It was found that it was difficult to predict with high accuracy. Therefore, when applying the above-described quality control method, the inventors of the present invention have a great influence on the amount of sag of the bumper due to the difference in the fastening direction to the body frame component (radiator support) to which the bumper is attached ( In other words, drawing elements data for when the bumper is fastened to the radiator support in the longitudinal direction of the vehicle body and when it is fastened to the vertical direction. In order to obtain a correlation with the amount of deformation when the bumper is mounted for each drawing element data after each layer. In addition, even if a relatively high correlation is shown between the drawing element data and the deformation amount, when the prediction formula is derived using this as a variable, the accuracy of the prediction formula may be insufficient. Paying attention to this, it is determined whether or not the positive and negative correlations showing relatively high values are consistent from the technical relationship between the corresponding drawing element and the amount of deformation at the time of attachment. Regression analysis was performed only for the discriminated drawing elements.

  Thus, the deformation prediction formula setting method according to the present invention is an improvement of the conventional quality control method according to the actual application target (resin bumper). According to this method, the conventional CAE analysis is performed. As compared with the prediction method using the method, a prediction formula capable of predicting a value closer to the actual amount of drooping can be set. Moreover, since the prediction formula is constructed based on the past design results, the prediction formula can be easily reviewed by accumulating (updating) the drawing element data and the deformation amount, and the prediction accuracy can be improved. . In addition, according to the present invention, since a prediction formula having excellent accuracy and reliability can be set with a simple processing operation using past drawing element data and deformation amount, the time required for derivation (calculation) can be reduced. The time required for the actual design can be shortened.

The solution of the technical problem is also achieved by the bumper deformation prediction method according to the present invention. That is, the deformation prediction method, the amount of deformation of the bumper occurring when fitted with a bumper made of a resin to the vehicle body, a modified prediction method of the resin bumper for predicting on the basis of drawing elements of the bumper, as FIG surface elements, Thickness of the fastening part with the radiator support, the sum of the longitudinal distance and vertical distance from the fastening part with the hood lock support to the dividing line with the hood, and from the fastening part with the hood lock support to the front end of the bumper Among the distances in the longitudinal direction, at least two elements are adopted, and a characteristic equation is characterized by deriving a prediction formula for the deformation amount at the time of attachment using these plural elements as variables. Alternatively, as FIG surface elements, the thickness of the fastening portion of the radiator support body longitudinal distance from the fastening portion of the hood lock support to the front end portion of the bumper, and the secant top end position of the fender to the front end portion of the bumper Among the distances in the longitudinal direction, at least two elements are adopted, and a characteristic equation is characterized by deriving a prediction formula for the deformation amount at the time of attachment using these plural elements as variables.

  The present inventors first selected a plurality of elements that are estimated to have a high correlation with the amount of sagging among the drawing elements of the resin bumper, based on empirical rules, and then sagging for each drawing element. As a result of obtaining the correlation with the amount, it has been found that the above-mentioned three drawing elements particularly affect the amount of sagging. In addition, at this time, the knowledge that by adopting at least two of the above three elements and deriving a prediction formula for the deformation amount at the time of attachment using these two elements as variables, it is possible to predict the amount of sag more accurately. I came to get. In addition, regarding the adoption of these drawing elements, it is possible to predict the amount of drooping with higher accuracy by changing the drawing elements to be adopted according to the manner in which the bumper is mounted on the vehicle body, specifically the fastening direction to the radiator support. It came to acquire the knowledge that becomes.

  The present invention has been made on the basis of the above knowledge, and by finding an optimal combination of drawing elements, it has become possible to derive an optimal prediction formula for predicting a deformation amount at the time of mounting a bumper. As a result, the amount of sag can be predicted in advance with high accuracy, and the number of rework steps and the like can be reduced. In addition, as described above, since the prediction formula is derived using at least two drawing elements as variables, it is necessary to change the dimensions as a result of the actual prediction, and it is handled as a variable of the prediction formula. Even if the value (dimension) of one drawing element of the bumper cannot be changed, it is possible to change other drawing elements that are both handled as variables. Thereby, even when various restrictions are imposed on the design change, the design change can be performed relatively freely.

Further, deformation prediction method of the bumper according to the present invention, the sum of the upper Symbol thickness of the fastening portion of the radiator support body longitudinal distance and vertical distance from the fastening portion of the hood lock support to the secant of the hood Of the longitudinal distance from the fastening portion with the hood lock support to the front end of the bumper , at least the sum of the longitudinal distance and the vertical distance from the fastening portion with the hood lock support to the secant portion with the hood is a variable. The prediction formula of the deformation amount at the time of attachment may be derived by the regression analysis including. Alternatively, the thickness of the fastening portion of the upper Symbol radiator support, the longitudinal direction of the car body longitudinal distance from the fastening portion of the hood lock support to the front end portion of the bumper, and the secant top end position of the fender to the front end portion of the bumper Of the distances , a prediction formula for the deformation amount during attachment may be derived by regression analysis including at least a longitudinal distance from the upper end position of the dividing line with the fender to the front end portion of the bumper.

A regression analysis was performed using two of the above three elements as variables, and a prediction formula for the amount of deformation when the bumper was installed was set. As described above, the bumper is in the longitudinal direction of the vehicle with respect to the radiator support. The most predictive formula is obtained by performing regression analysis that includes the sum of the distance in the longitudinal direction and the distance in the vertical direction from the fastening part with the hood lock support to the secant part with the hood as variables. It turned out to be preferable in terms of accuracy. Similarly, when the bumper is fastened to the radiator support in the vertical direction , it is obtained by performing a regression analysis that includes at least the longitudinal distance from the upper end position of the dividing line with the fender to the front end of the bumper as a variable. It was found that the prediction formula is most suitable in terms of prediction accuracy. Therefore, by performing regression analysis using the above-described drawing elements as variables, it is possible to predict the amount of deformation at the time of installation, which is excellent in both accuracy and reliability.

  As described above, according to the present invention, a technique for predicting the deformation amount of the bumper when mounted on the vehicle body, which is excellent in reliability and certainty, can be constructed, thereby shortening the time required for the final design.

It is a flowchart of the setting method of the deformation | transformation prediction formula of the resin bumper which concerns on one Embodiment of this invention. It is a longitudinal cross-sectional view in the vehicle width direction center of the resin bumper when the fastening direction with the radiator support is the vehicle body longitudinal direction. It is a longitudinal cross-sectional view in the vehicle width direction center of the resin bumper when the fastening direction with the radiator support is the vertical direction. It is a principal part expansion perspective view of a resin bumper. It is an example of the graph which shows the correlation with the drawing element of resin bumpers, and the deformation amount at the time of vehicle body attachment. It is another example of the graph which shows the correlation with the drawing element of resin bumpers, and the deformation amount at the time of vehicle body attachment.

  Hereinafter, an embodiment of a method for setting a deformation prediction formula for a resin bumper and a deformation prediction method for a resin bumper according to the present invention will be described with reference to the drawings. In this embodiment, a case will be described as an example in which the amount of sag of a bumper with respect to the vehicle body that occurs when the resin bumper is attached to the vehicle body is predicted.

  FIG. 1 shows a flowchart of a method for setting a deformation prediction formula for a resin bumper according to an embodiment of the present invention. As shown in the figure, this prediction formula setting method is based on the fact that bumper drawing element data for a plurality of vehicle models designed in the past is fastened in the vertical direction when the bumper is fastened to the radiator support in the longitudinal direction of the vehicle body. And a correlation acquisition step S2 for obtaining a correlation between the sag amount generated when the bumper is attached for each drawing element data grouped by stratification, and a relatively high value. A technical alignment determination step S3 for determining whether or not the sign of the correlation shown is consistent in view of the technical relationship between the corresponding drawing element and the amount of sag, and the drawing element determined to be consistent The regression analysis step S4 for setting a prediction formula for the amount of drooping with the drawing element subjected to the regression analysis as a variable is performed by performing the above regression analysis only for. In addition, a drawing element selection step S0 that adopts a plurality of drawing elements used in at least the stratification step S1 and the correlation acquisition step S2 is arranged before the stratification step S1. Hereinafter, each process will be described in detail.

(Drawing element selection process S0)
First, a plurality of bumper drawing elements that are presumed to have a high correlation with the amount of bumper drooping are selected based on the empirical rules of the operator (engineer). Specifically, as shown in FIGS. 2 to 4, the longitudinal distance (so-called overhang amount) and the plate thickness (the so-called overhang amount) from the fastening portion 10 a with the radiator support 11 to the dividing line portion 10 b with the hood 12 in the bumper 10. For example, the plate thickness of the fastening part 10a with the radiator support 11 (dimension A in FIG. 2 and dimension D in FIG. 3), the distance in the vehicle body longitudinal direction from the fastening part 10c with the hood lock support 13 to the front end 10d of the bumper 10 ( The dimension C in FIG. 2 and the dimension E in FIG. 3), the longitudinal distance or the vertical distance from the fastening part 10c with the hood lock support 13 to the dividing line part 10b with the hood 12 (dimension b1 or b2 in FIG. 2). , The number of fastening points of the fastening part 10a with the radiator support 11, the pitch between the fastening parts 10a, and the front end part of the bumper 10 from the upper end position of the dividing line part 10e with the fender 14 Can be exemplified such as longitudinal distance to 0d (dimensions in FIG. 4 F). Further, examples of combinations of these drawing elements include dimensions B, B / A, and F / D as the sum of dimensions b1 and b2. Of course, you may provide elements other than the said illustration to the process after layering process S1 as drawing element. The same applies to combinations of drawing elements.

(Layered process S1)
Next, with respect to the drawing elements adopted in step S0, the drawing element data of the bumper 10 in a plurality of vehicle types designed in the past are prepared, and the prepared drawing element data is sent from the bumper 10 to the radiator support 11 in the longitudinal direction of the vehicle body. It is divided into a case where it is fastened and a case where it is fastened in the vertical direction. Hereinafter, the process which concerns on each process S2-S4 is given for every drawing element data group classified by layer.

(Correlation acquisition step S2)
For each drawing element data grouped in the stratification step S1, a correlation between the drawing element and the amount of sag of the bumper 10 is obtained. At this time, not only the strength of the correlation (absolute value) is obtained, but also the positive / negative is obtained (the corresponding information is left so that it can be used in the subsequent steps). In this embodiment, in addition to the drawing elements exemplified in the description section of the drawing element selection step S0, a combination of the exemplified drawing elements (dimensions B, B / A, F / D, etc.) is also a bumper at the time of mounting the vehicle body. The correlation with the amount of droop of 10 is obtained.

(Technical consistency determination step S3)
In this way, when the correlation between the predetermined drawing element or a combination thereof and the amount of sag of the bumper 10 is obtained, the positive or negative of the correlation with respect to a plurality of upper drawing elements showing a relatively high correlation corresponds to the corresponding drawing. It is determined whether or not they are consistent from the technical relationship between the element and the amount of sag during installation.

  Here, for example, as shown in FIG. 5, the negative correlation (the regression line has a negative slope) indicates that the corresponding drawing element, here, the fastening portion 10 c with the hood lock support 13 to the hood 12. It is determined whether or not they are matched in view of the technical relationship between the dimension B, which is the sum of the vehicle body longitudinal direction distance b1 and the vertical direction distance b2 up to the dividing line portion 10b, and the amount of sag. In this case, it is reasonable in view of technical common sense in the field that the amount of sagging increases as both the vehicle body longitudinal dimension b1 and the vertical distance b2 increase, and the actual correlation also shows a tendency according to this. Therefore, a negative correlation value is determined to be technically consistent.

  On the other hand, for example, as shown in FIG. 6, when the corresponding drawing element is the number of fastening points of the fastening portion 10 a with the radiator support 11, the bumper 10 hangs down as the number of fastening points increases according to common technical knowledge. Where the amount should be decreased, when the actual correlation result is seen, the amount of sag of the bumper 10 increases as the number of fastening points increases. From this, it is difficult to say that the negative correlation value is consistent from the technical relationship between the corresponding drawing element (number of fastening points) and the amount of sag (the consistency is denied) ). In this case, the corresponding drawing element is excluded from processing in the following step S3.

  As described above, only high-order drawing elements that exhibit a relatively high correlation and that have been determined to have the above technical consistency are provided as the targets of the regression analysis related to the next process. Speaking of the above-illustrated drawing elements or combinations thereof, dimensions B, C, E, F and combinations thereof B / A, F / D, etc. may be mentioned.

(Regression analysis step S4)
In step S3, the regression analysis is performed only for the drawing elements or combinations thereof determined to be technically consistent. As a result, a prediction formula for the amount of drooping with the drawing element or a combination thereof as a variable is derived. When the drawing elements exemplified above or combinations thereof are obtained, the amount of sag of the bumper 10 is derived by the following regression equation.
Formula 1: (Bumper sagging amount) = m0−m1 × (B / A)
Formula 2: (Bumper sag) = n0−n1 × B−n2 × C
Formula 3: (Bumper sagging amount) = s0−s1 × (F / D)
Formula 4: (Bumper sagging amount) = t0−t1 × F−t2 × E
However, Formula 1 or Formula 2 is applied when the fastening direction of the bumper 10 to the radiator support 11 is the longitudinal direction of the vehicle body, and Formula 3 or Formula 4 is when the fastening direction to the radiator support 11 is the vertical direction. Applied. Further, m0, m1, n0, n1, n2, s0, s1, t0, t1, and t2 all indicate positive values. Here, if the amount of sag of the bumper shows a negative value, it means that it actually hangs down, and if it shows a positive value, it means that the lower end of the bumper is positioned vertically above the design dimension. It shall be.

Thus, by calculating the regression formula (regression line), a prediction formula for the amount of drooping can be derived. By inputting drawing element data (design dimensions) corresponding to each variable, the drawing can be obtained. The amount of sag of the bumper 10 according to the above can be predicted. For example, in the case of the above formulas 1 to 4 , if the absolute value of the obtained value is within a predetermined range, it can be determined that it is allowed.

  At this time, in the present invention, the drawing element data is stratified according to whether the bumper 10 is fastened to the radiator support 11 in the longitudinal direction of the vehicle body or when it is fastened in the vertical direction. Correlation with the amount of sag of the bumper 10 is obtained for each element data, and the sign of the correlation showing a relatively high value is consistent from the technical relationship between the corresponding drawing element and the amount of sag. The regression analysis is performed only for the drawing elements that are determined to be consistent. Thereby, the conventional quality control method can be appropriately improved according to the actual application target, and a deformation prediction formula suitable for the drooping prediction of the bumper 10 can be set. Therefore, according to this method, it is possible to set a prediction formula that can predict a value closer to the actual amount of drooping compared with a prediction method using a conventional CAE analysis. Moreover, since the prediction formula is constructed based on the past design results, the prediction formula can be easily reviewed by accumulating (updating) the drawing element data and the deformation amount, and the prediction accuracy can be improved. . In addition, according to the present invention, since a prediction formula having excellent accuracy and reliability can be set with a simple processing operation using past drawing element data and deformation amount, the time required for derivation (calculation) can be reduced. The time required for the actual design can be shortened.

In the present invention, as shown in FIG. 2, when the bumper 10 is fastened to the radiator support 11 in the longitudinal direction of the vehicle body , as a drawing element, the plate thickness A of the fastening portion 10a with the radiator support 11, the hood lock The sum (B) of the vehicle body longitudinal distance b1 and the vertical distance b2 from the fastening part 10c to the support 13 to the dividing line part 10b to the hood 12, and the front end part 10d of the bumper 10 from the fastening part 10c to the hood lock support 13 At least two elements in the longitudinal distance C are adopted, and a prediction formula for the amount of drooping with these two elements as variables is derived. Alternatively, as shown in FIG. 3, when the bumper 10 is fastened to the radiator support 11 in the vertical direction, the plate thickness D of the fastening portion 10 a with the radiator support 11 and the hood lock support 13 are fastened as the drawing elements. The vehicle body longitudinal direction distance E from the portion 10c to the front end portion 10d of the bumper 10 and the longitudinal direction distance F from the upper end position of the dividing line portion 10e to the fender 14 to the front end portion 10d of the bumper 10 are adopted. A prediction formula for the amount of sagging with two elements as variables was derived. Thus, by finding an optimal combination of drawing elements for predicting the amount of sag of the bumper 10, it is possible to derive a prediction formula that can predict the amount of sag with high accuracy. This makes it possible to predict the amount of sag from the initial prediction with high accuracy, thereby reducing the number of rework steps and shortening the time required for the final design. Moreover, when it is necessary to change the dimensions as a result of the actual prediction, and the value of one drawing element of the bumper 10 handled as a variable of the prediction formula (for example, the size of the thickness A) cannot be changed. Can be dealt with by changing other drawing elements (for example, the size of the dimension B) that are both handled as variables. Therefore, even when various restrictions are imposed on the design change, the design change can be performed relatively freely by selecting and using an appropriate variable or expression according to the contents of the restriction.

  In particular, since the above-illustrated drawing elements (A to F) or combinations thereof (B / A, F / D) remain without being excluded even in the technical matching determination step S3 regarding the positive or negative of the correlation, It can be said that it is high not only in reliability (prediction accuracy) but also in reliability. Therefore, it is highly preferable to set a droop amount prediction formula using these exemplary drawing elements or combinations thereof as variables.

10 Bumper 10a Fastening part 10b with radiator support Split part 10c with hood Fastening part 10d with hood lock support Front end part 10e Split line part with fender 11 Radiator support 12 Hood 13 Hood lock support 14 Fenders A and D with radiator support Thickness b1 at fastening part Vehicle body longitudinal distance b2 from hood lock support fastening part to hood split line part Vertical distance C from hood lock support fastening part to hood split line part, E From hood lock support fastening part to bumper front end part Vehicle body longitudinal distance F Longitudinal distance S0 from the upper end position of the fender dividing line part to the front end part of the bumper Drawing element selection process S1 Stratification process S2 Correlation acquisition process S3 Technical matching determination process S4 Regression analysis process

Claims (5)

A method of setting a prediction formula for predicting a deformation amount of the bumper generated when a resin bumper is attached to a vehicle body,
A correlation acquisition step for obtaining a correlation between the bumper drawing element data and the amount of deformation at the time of installation in a plurality of vehicle models designed in the past, and a regression analysis for performing regression analysis on the plurality of drawing elements having a relatively high correlation And when setting a prediction formula for the amount of deformation at the time of installation using the drawing element as a variable by the regression analysis,
The drawing element data is stratified according to whether the bumper is fastened to the radiator support in the longitudinal direction of the vehicle body or when the bumper is fastened in the vertical direction, and the deformation amount at the time of attachment for each of the drawing element data. As well as
It is determined whether or not the positive and negative of the correlation indicating a relatively high value is consistent from the technical relationship between the corresponding drawing element and the deformation amount at the time of attachment, and is determined to be consistent. A method for setting a deformation prediction formula for a resin bumper that performs the regression analysis only on the drawing elements. A deformation prediction method for a resin bumper for predicting a deformation amount of the bumper generated when the resin bumper is attached to a vehicle body based on a drawing element of the bumper ,
As before Symbol drawing elements, the thickness of the fastening portion of the radiator support, the sum of the vehicle body longitudinal distance and vertical distance from the fastening portion of the hood lock support to the secant of the hood, as well as engagement of the hood lock support A resin bumper deformation prediction method that adopts at least two elements in the longitudinal distance from the front part to the front end of the bumper, and derives a prediction equation for the deformation amount at the time of attachment using the plurality of elements as variables. The plate thickness of the fastening portion with the radiator support, the sum of the vehicle body longitudinal distance and the vertical distance from the fastening portion with the hood lock support to the secant portion with the hood, and the fastening portion with the hood lock support Among the longitudinal distances to the front end of the bumper , at the time of installation, by regression analysis including at least the sum of the longitudinal distance and the vertical distance from the fastening part with the hood lock support to the dividing line part with the hood as a variable The deformation prediction method for a resin bumper according to claim 2, wherein a prediction formula for the deformation amount is derived. A deformation prediction method for a resin bumper for predicting a deformation amount of the bumper generated when the resin bumper is attached to a vehicle body based on a drawing element of the bumper ,
As before Symbol drawing elements, the thickness of the fastening portion of the radiator support body longitudinal distance from the fastening portion of the hood lock support to the front end portion of the bumper, and the front end portion of the bumper from the secant top end position of the fender A method for predicting deformation of a resin bumper, wherein at least two elements of the distance in the longitudinal direction are adopted, and a prediction formula of the deformation amount at the time of attachment is derived using these plural elements as variables. The plate thickness of the fastening portion with the radiator support, the longitudinal distance of the vehicle body from the fastening portion with the hood lock support to the front end portion of the bumper, and the length from the upper end position of the dividing line with the fender to the front end portion of the bumper The prediction formula of the amount of deformation at the time of attachment is derived by regression analysis including, as a variable, at least a longitudinal distance from the upper end position of the dividing line with the fender to the front end of the bumper among the directional distances. Deformation prediction method for resin bumpers.

Images