JP5583929B2 - Orientation state prediction method and deformation behavior analysis method - Google Patents

Description

  The present invention relates to a method for predicting the orientation state of a filler in a resin molded product containing the filler and a method for analyzing the deformation behavior of the resin molded product.

  Conventionally, attempts to substitute tests such as mechanical strength tests of various resin products by numerical analysis methods such as the finite element method for the purpose of shortening the design period of resin products (resin molded products) and reducing prototype costs, It is adopted in the design site of various resin products.

  Problems when the above numerical analysis method is not used include that the mechanical strength test requires the preparation of trial products for the number of tests if the test involves the destruction of resin products and their prototypes. There are cases where the test cannot be re-executed due to time constraints in design. For this reason, in the design site of a resin product, it is an important issue to substitute a test such as a mechanical strength test of the resin product by a numerical analysis method such as a finite element method.

  As a numerical analysis method, a method using a finite element method as described in Non-Patent Document 1 is generalized. Hibbitt, Karlsson & Sorensen Inc. is commercially available software for predicting tests such as mechanical strength tests of resin products using the finite element method. Product names ABAQUS (registered trademark) and ADINA R & D, Inc. The product name ADINA (registered trademark) and the like are known.

  However, most of the methods using commercially available software as described above assume that the mechanical characteristics of the resin product are isotropic and linear. As a result, in the method using commercially available software as described above, the test result and the prediction result in an actual resin product are unlikely to coincide. As a case where the actual test result and the simulation result are difficult to coincide with each other, for example, there is a case where the test result of a resin product manufactured using a resin composition containing a fibrous filler is predicted.

  In a resin product produced using a resin composition containing a filler such as a fibrous filler, it is necessary to consider the orientation of the filler such as a fibrous filler. As a method for considering the orientation of the filler, methods described in Patent Documents 1 and 2 are known. The methods described in Patent Documents 1 and 2 first predict the fiber orientation state in the resin product by an injection molding process simulation, and then give the predicted fiber orientation state as a physical property value of each element. It is a method to do.

  The problems with the methods described in Patent Documents 1 and 2 are that first, it is necessary to separately calculate the physical property values of the individual elements, and it takes a lot of time to calculate, and secondly, the orientation state of the filler. The flow analysis software is necessary for the prediction and the calculation cost is large. Thirdly, since the prediction accuracy of the fiber orientation state is insufficient at present, even if these methods are used, the test results in the actual resin product and It may be mentioned that the prediction result may not match.

  In particular, the methods described in Patent Documents 1 and 2 assume a Hele-Show flow for resin flow, and determine the orientation angle and degree of orientation of a filler such as a fibrous filler using the Folger-Tucker equation. Yes. Therefore, when the resin flow is contrary to the above assumption, the prediction accuracy is deteriorated. As a case where the flow of the resin is contrary to the above assumption, there is a case where the weld portion which is a non-fountain flow region in the resin product is evaluated. As described above, although the prediction accuracy deteriorates in the weld portion, there is no alternative method. For this reason, there is a need for a method for predicting the mechanical strength and the like of a resin molded product with high accuracy even in a portion (for example, a weld portion) contrary to the above assumption.

  Further, in the conventional methods described in Patent Documents 1 and 2, etc., Tandon-Weng, Mori is used for predicting the mechanical strength of the resin product performed after predicting the orientation state of the filler in the resin molded product. -Micromechanics theory such as Tanaka model is used. Since these theories do not assume a three-dimensional and micro structure, this is one of the causes of the deterioration of the prediction accuracy as a result.

JP 2006-272928 A JP 2005-283539 A

R. T Fenner, “Practical Finite Element Method” Science, 1980

  The present invention has been made in order to solve the above-described problems, and its purpose is simple and time-consuming when predicting the orientation state of the filler in the resin molded product, An orientation state prediction method capable of predicting the orientation state of the filler in the resin molded product with high accuracy, and the deformation behavior of the resin molded product can be analyzed accurately, and the flow of the resin in the resin molded product is a Hele-Show flow. It is another object of the present invention to provide a deformation behavior analysis method for analyzing a deformation behavior of a resin molded product with high accuracy even when analyzing a portion other than the portion represented by.

  The inventors of the present invention have made extensive studies to solve the above problems. As a result, a slice image acquisition step of acquiring a plurality of slice images at predetermined intervals in a predetermined direction for at least a part of a resin molded product formed by molding a resin composition containing a filler at a predetermined ratio, and the slice A division step of dividing an image into pixels of a minute area, an image density determination step of determining an image density of each pixel divided by the division step, and selecting one or more slice images from the plurality of slice images, A threshold value determining step for determining an image density threshold value for discriminating between the filler and the resin portion so that the ratio of the filler in the selected slice image becomes the predetermined ratio; and the image density threshold value and each pixel An image conversion process for converting a plurality of slice images into a plurality of binarized images based on the pixel density of the image, and a plurality of binarized images are sequentially stacked to obtain a slice image of a resin molded product. If the orientation state prediction method predicts the orientation state of the filler in the resin molded product having a reconstruction step for reconstructing a three-dimensional model of the part, the present invention is completed by finding that the above problems can be solved. It came to. More specifically, the present invention provides the following.

  (1) A slice image acquisition step of acquiring a plurality of slice images at predetermined intervals in a predetermined direction for at least a part of a resin molded product formed by molding a resin composition containing a filler at a predetermined ratio, and the slice A division step of dividing an image into pixels of a small area, an image density determination step of determining an image density of each pixel divided by the division step, and one or more slice images are selected from the plurality of slice images A threshold value determining step for determining an image density threshold value for discriminating between the filler and the resin portion so that the ratio of the filler becomes the predetermined ratio in the selected slice image; and An image conversion step of converting the plurality of slice images into a plurality of binarized images based on the pixel density of each pixel, and the resin molding by sequentially stacking the plurality of binarized images Orientation state prediction method for predicting the a reconstruction step of reconstructing a three-dimensional model of the slice image acquisition part, the orientation of the filler in the resin molded article having.

(2) The threshold value determination step arbitrarily selects two or more slice images from the slice image, and the filling is performed so that the ratio of the filler is the predetermined ratio in each of the selected slice images. Determining the image density threshold value for discriminating between the agent and the resin part, and calculating the average value of the image density threshold value by averaging each image density threshold value,
The orientation conversion prediction according to (1), wherein the image conversion step is a step of converting the plurality of slice images into a plurality of binarized images based on an average value of the image density threshold values and a pixel density of each pixel. Method.

  (3) The orientation state prediction method according to (1) or (2), wherein the slice image is an X-ray CT image.

  (4) The resin molded product is a resin molded product having a non-fountain flow region, and the slice image acquisition step includes a plurality of portions when a portion including the non-fountain flow region is sliced at a predetermined interval in a predetermined direction. The orientation state prediction method according to any one of (1) to (3), which is a step of acquiring a slice image.

  (5) The orientation state prediction method according to (4), wherein the non-fountain flow region is a weld portion.

  (6) A deformation behavior analysis method for analyzing a deformation behavior of a resin molded product using a three-dimensional model obtained from the orientation state prediction method according to any one of (1) to (5).

  (7) A slice image acquisition step of acquiring a plurality of slice images at predetermined intervals in a predetermined direction for at least a part of a resin molded product formed by molding a resin composition containing a filler at a predetermined ratio, and the slice A division step of dividing an image into pixels of a small area, an image density determination step of determining an image density of each pixel divided by the division step, and one or more slice images are selected from the plurality of slice images A threshold value determining step for determining an image density threshold value for discriminating between the filler and the resin portion so that the ratio of the filler becomes the predetermined ratio in the selected slice image; and A finite element model creating step for creating at least a part of the finite element model, and an element model image density determining step for determining the respective image densities of the finite element model A finite element model reconstruction step for reconfiguring the finite element model into a finite element model obtained by binarizing the finite element model based on the image density threshold and the pixel density of each finite element model, and the binarized finite element model. Deformation behavior analysis method that uses it to analyze the deformation behavior of resin molded products.

  (8) The threshold value determination step arbitrarily selects two or more slice images from the slice image, and the filling is performed so that the ratio of the filler is the predetermined ratio in each of the selected slice images. Determining an image density threshold value for discriminating between the agent and the resin portion, and averaging the image density threshold values to obtain an average value of the image density threshold values, and the finite element model reconstruction step includes the image density The deformation behavior analysis method according to (7), which is a step of reconfiguring the finite element model into a binarized finite element model based on the average value of the threshold and the pixel density of each finite element model.

  According to the present invention, it is possible to predict the orientation state of the filler in the resin molded product with high accuracy while predicting the orientation state of the filler in the resin molded product without taking time and cost. It is possible to analyze the deformation behavior of the resin molded product with high accuracy. Furthermore, it is possible to analyze the flow of the resin in the resin molded product with high accuracy when predicting a portion other than the portion represented by the Hele-Show flow.

It is a figure which shows the fountain flow. It is a schematic diagram of an X-ray CT apparatus. (A) is a figure which shows the slice image image | photographed with the X-ray CT apparatus. (B) is a figure which shows the binarized slice image. It is a flowchart which shows a threshold value determination process. It is a figure which shows an example of the three-dimensional model obtained from the orientation state prediction method of this invention. It is a figure which shows the model shape of the flat plate used by Example 1, the comparative example 1, and measurement 1. FIG. It is a figure which shows the position in the resin molded product of the resin test piece used in Example 1. FIG. FIG. 3 is a diagram illustrating a binarized finite element model according to the first embodiment. It is a figure which shows the model shape of the flat plate used by Example 2, the comparative example 2, and measurement 2. FIG. It is a figure which shows the position in the resin molded product of the resin test piece used in Example 2. FIG. 6 is a diagram illustrating a binarized finite element model according to Embodiment 2. FIG.

  Hereinafter, embodiments of the present invention will be described in detail. The present invention is not limited to the invention described below.

<Orientation state prediction method>
The alignment state prediction method of the present invention is a slice image that acquires a plurality of slice images at predetermined intervals in a predetermined direction for at least a part of a resin molded product formed by molding a resin composition containing a filler in a predetermined ratio. An acquisition step, a division step of dividing a slice image into pixels of a minute area, an image density determination step of determining an image density of each pixel divided by the division step, and one or more slice images from a plurality of slice images And a threshold value determining step for determining an image density threshold value for discriminating between the filler and the resin portion so that the ratio of the filler in the selected slice image becomes the predetermined ratio, An image conversion process for converting a plurality of slice images into a plurality of binarized images based on the pixel density of each pixel, and a slice image of a resin molded product by sequentially stacking a plurality of binarized images A reconstruction step of reconstructing a three-dimensional model of the resulting part, and having a. Hereinafter, each process of the orientation state prediction method of the present invention will be described.

[Slice image acquisition process]
The slice image acquisition step is a step of acquiring a plurality of slice images at predetermined intervals in a predetermined direction in at least a part of a resin molded product formed by molding a resin composition containing a filler in a predetermined ratio. .

  First, a resin molded product for which a slice image is to be acquired is determined. The resin molded product to be predicted of the present invention is formed by molding a resin composition containing a filler in a predetermined ratio.

  The alignment state prediction method of the present invention can target various conventionally known resins. A resin mixture obtained by blending a plurality of resins is also included in the resin composition.

The resin composition contains a filler at a predetermined ratio. The type of filler is preferably an inorganic filler that clearly shows the boundary between the resin and the filler and has a low X-ray transmittance. Examples of conventionally known inorganic fillers include fibrous fillers, granular fillers, and plate-like fillers.
Preferred fibrous fillers include, for example, glass fiber, asbestos fiber, silica fiber, silica / alumina fiber, alumina fiber, zirconia fiber, boron nitride fiber, silicon nitride fiber, boron fiber, potassium titanate fiber, stainless steel, aluminum, Examples thereof include inorganic fibrous materials such as metallic fibrous materials such as titanium, copper, and brass.
In addition, as the particulate filler, silica, quartz powder, glass beads, milled glass fiber, glass balloon, glass powder, calcium silicate, aluminum silicate, kaolin, talc, clay, diatomaceous earth, wollastonite, oxidation Metal oxides such as iron, titanium oxide, zinc oxide, antimony trioxide, and alumina, carbonates of metals such as calcium carbonate and magnesium carbonate, sulfates of metals such as calcium sulfate and barium sulfate, other ferrites, silicon carbide, Examples thereof include silicon nitride, boron nitride and various metal powders.
Examples of the plate-like filler include mica, glass flakes, various metal foils and the like.

  Among the above fillers, the orientation state prediction method of the present invention and the deformation behavior analysis method described later can be preferably applied to a resin molded product formed by molding a resin composition containing a fibrous filler. A resin molded product containing a fibrous filler has a large anisotropy such as physical properties depending on the orientation state of the filler. As a result, unless the orientation state of the filler is predicted with higher accuracy, it becomes difficult to match the predicted result with the actual result when analyzing the deformation behavior of the resin molded product described later. Moreover, the orientation state prediction method of the present invention and the deformation behavior analysis method described later can be preferably applied to a resin molded product formed by molding a resin composition containing glass fibers among the fibrous fillers.

  In addition to the resin composition, additives such as a nucleating agent, a colorant, an antioxidant, a stabilizer, a plasticizer, a lubricant, a mold release agent, and a flame retardant are added to give a resin composition having desired characteristics. included.

  The resin molded product can be obtained by a conventionally known molding method. Examples of conventionally known molding methods include various molding methods such as compression molding, transfer molding, injection molding, extrusion molding, and blow molding.

  In this step, a plurality of slice images are acquired at predetermined intervals in a predetermined direction for at least a part of the resin molded product obtained as described above. "At least a part" means that the entire obtained resin molded product may be a target for predicting the orientation state of the filler, or a part of the resin molded product may be a target for predicting the orientation state of the filler. Point to. For example, when the orientation state in the resin molded product is uniform, if the orientation state is predicted only for a part, it can be predicted that the other portions are in the same orientation state as the predicted part. Therefore, even if the orientation state is predicted from only a part of the resin molded product, it can be preferably used for the analysis of the deformation behavior of the entire molded product. The orientation state prediction method of the present invention can predict the orientation state of the filler in the resin molded product with extremely high accuracy.

  Further, according to the orientation state prediction method of the present invention, it is possible to predict the orientation state of the non-fountain flow region of a resin molded product that has been conventionally difficult to predict. As shown in FIG. 1, the “fountain flow region” means that the molten resin flows in such a manner that the molten resin does not flow in the mold due to sliding with the mold surface but is ejected from the center of the flow to the mold surface. Refers to the part. The “non-fountain flow region” refers to a portion where the molten resin does not flow in the mold as described above, and includes, for example, a weld portion formed by joining the flowing molten resin. Conventionally, it is also possible to predict the orientation state of the weld portion, which has been difficult to predict, by a method described later. In this case, a slice image at the weld portion is acquired. One of the features of the orientation state prediction method of the present invention is that the prediction accuracy of the orientation state of the filler at the weld in the resin molded product is improved.

  When a part of a resin molded product is a target for obtaining a slice image, a resin test piece for obtaining a slice image by cutting out the resin molded product from the resin molded product for which a slice image is to be acquired as necessary. Is made. When the entire resin molded product is a target for obtaining a slice image, the entire resin molded product is a resin test piece.

  Although the acquisition method of a slice image is not specifically limited, it can acquire using the X-ray CT apparatus 1 as shown in FIG. The X-ray CT apparatus 1 includes an X-ray irradiation unit 11 for irradiating the resin test piece 2 with X-rays, an X-ray detection unit 12 that detects X-rays transmitted through the resin test piece 2 as projection data, and a resin test. A sample stage 13 for holding the piece 2, a rotation drive unit 14 for moving the sample stage 13 up and down (moving in the direction of the arrow in FIG. 2) and rotating (moving in the direction of the white arrow in FIG. 2); An image processing unit 15 for reconstructing projection data in a plurality of angular directions as slice images.

  The X-ray irradiation unit 11 is a part for irradiating the resin test piece 2 with X-rays. It will not specifically limit if it can irradiate X-ray | X_line, A conventionally well-known X-ray irradiation apparatus can be used. For example, an X-ray tube etc. are mentioned. In the X-ray irradiation part 11, the irradiation conditions of the X-rays irradiated to the resin test piece 2 can be adjusted. Examples of X-ray irradiation conditions include tube current and X-ray irradiation time. In the orientation state prediction method of the present invention, the X-ray irradiation conditions are not particularly limited, and can be appropriately changed according to the shape of the resin test piece to be predicted, the type of resin included, and the like.

  The X-ray detection unit 12 is a part that detects X-rays transmitted through the resin test piece 2 as projection data after converting the X-rays into electrical signals. The X-ray detection unit 12 is disposed so as to face the X-ray irradiation unit 11 with the resin test piece 2 interposed therebetween.

  The sample stage 13 is a part for holding the resin test piece 2 so that the resin test piece 2 is irradiated with X-rays. The sample stage is disposed between the X-ray irradiation unit 11 and the X-ray detection unit 12.

  The rotation driving unit 14 is a part for causing the sample stage 13 to move up and down and rotate so that the resin test piece 2 is irradiated with X-rays from a plurality of angular directions. The rotation drive unit 14 is connected to the sample stage 13. The rotational drive unit 14 can irradiate various positions in the resin test piece 2 from a plurality of directions. As a result, projection data can be obtained for X-rays transmitted through the resin test piece 2 from various angles.

  The image processing unit 15 is a part that reconstructs projection data in a plurality of angular directions as slice images. The image processing unit 15 is connected to the X-ray detection unit 12. The projection data detected by the X-ray detection unit 12 is sent to the image processing unit 15, and a slice image is obtained by performing conventionally known image processing. The conventionally known image processing method is, for example, a method in which projection data in each direction is subjected to one-dimensional Fourier transform, these are combined to create a two-dimensional Fourier transform image, and this is subjected to inverse Fourier transform to obtain a reconstructed image. Is mentioned.

  The slice images are slice images at a predetermined interval in a predetermined direction in the resin molded product. Therefore, a plurality of slice images are obtained in this step. The “predetermined interval” is a range in which shading is averaged when obtaining a slice image. The interval between the slice images is not particularly limited and can be appropriately changed depending on the measurement target. However, in order to appropriately predict the orientation state of the filler in the resin molded product including the fiber filler, the slice image The interval between the images is preferably equal to or less than the average diameter of the filler, or 20 μm or less when the average diameter is unknown. The “predetermined direction” is a direction perpendicular to the image plane of the slice image, and the predetermined direction can be set to a desired direction.

  An example of the slice image obtained as described above is shown in FIG. FIG. 3A shows a monotone slice image. The blacker portion is a portion that easily transmits X-rays in the resin molded product, and the whiter portion is a portion that hardly transmits X-rays in the resin molded product. The slice image includes a resin portion and a filler portion. Usually, the resin portion is more likely to transmit X-rays than the filler portion. Therefore, in FIG. 3A, the black portion tends to represent the resin, and the white portion tends to represent the filler. Since the slice image includes a noise pattern as described below, it cannot be said that the portion represented simply in black is a resin portion and the portion represented in white is a filler portion.

  Since the resin portion and the filler portion have different X-ray absorption rates, a dark and light image can be obtained. Furthermore, when obtaining a slice image of the resin test piece 2 composed of a plurality of materials having different X-ray absorption rates in this way, the X-ray absorption rate is discontinuous across the boundary between the resin portion and the filler portion. To change. For this reason, discontinuous (abrupt) changes also appear in the projection data obtained by the X-ray detector 12. As a result, high-frequency components due to discontinuous changes appear greatly in the one-dimensional Fourier transform image. Due to numerical calculation errors caused by the high frequency components, virtual images (noise patterns) called artifacts appear in the reconstructed image obtained from the image processing unit 15. This noise pattern is also included in the slice image. This noise pattern is a major obstacle to predicting the orientation state of the filler using the slice image. In the conventional method, it is difficult to predict an appropriate orientation state of the filler. If it is the method of this invention, the orientation state of a suitable filler can be estimated.

[Division process]
In the dividing step, the slice image is divided into pixels having a small area. This step is performed for all of the two or more slice images obtained in the slice image acquisition step. The size of one pixel is not particularly limited as long as the color state and density distribution of the slice image can be grasped. In addition, when the slice image is divided into pixels of a minute area that can grasp the color density distribution at the time of acquiring the slice image, the original slice image pixel is directly used as a divided pixel below. It can be used in the process.

[Image density determination process]
The image density determining step is a step of determining the image density of each pixel divided by the dividing step. The image density can be determined by measuring by a conventionally known method. As described above, the slice image includes a resin portion, a filler portion, and a noise portion. The concentration is determined for all of these. For example, the image density of a pixel can be represented by the number of gradations for the brightness of the image. The number of gradations in the case where the image density is expressed by the number of gradations is not particularly limited as long as the number of gradations is sufficient to grasp the color state and density distribution of the slice image. In FIG. 3A, a 256-tone monotone slice image is shown.

  When the image density in one pixel is light and shaded, the method for determining the image density of the pixel is not particularly limited. For example, the image density of the pixel is determined to be the image density of the large area in the pixel. Alternatively, the average of the image density in the pixel may be taken.

[Threshold determination step]
The threshold value determining step refers to selecting one or more slice images from the plurality of slice images, and the ratio of the filler in the selected slice image is the predetermined ratio (the ratio of the filler contained in the resin molded product). This is a step of determining an image density threshold for discriminating between the filler and the resin portion. That is, in the image density obtained in the above-described image density determination step, the degree of the darkness (how much the black part) is predicted as the resin part, and how thin the part (how much the white part) This is a step of determining an image density threshold value for predicting up to the filler portion.

  In this step, first, a slice image for determining a threshold value is selected. The slice image to be selected is one or more slice images. First, a method for determining a threshold value from one slice image will be specifically described with reference to the flowchart of FIG. Note that the threshold determination method is not limited to the following method.

  The slice image is an image as shown in FIG. 3A as described above, and includes a white portion, a black portion, and a gray portion. This slice image is obtained by expressing the image density in the image density determination step by the number of gradations (256 gradations) for brightness. It is divided into a portion to be predicted as a resin portion and a portion to be predicted as a filler portion with the number of gradations serving as a threshold as a boundary.

  First, as shown in FIG. 4, a predetermined number of gradations is set as a threshold value (step 1). Next, the slice image is binarized from the threshold set in step 1 (step 2). It is calculated how much the resin part and the filler part are included in the binarized slice image (step 3). Compare the content ratio of the resin part or filler part obtained in step 3 with the ratio of the filler in the actual resin molded product (predetermined ratio) or the ratio of resin, and the threshold set in step 1 is appropriate (Step 4). If the threshold value set in Step 1 can be determined as an appropriate threshold value from the result of Step 4, the threshold value set in Step 1 is determined as an image density threshold value for discriminating between the resin portion and the filler portion. If it cannot be determined that the threshold set in step 1 is an appropriate threshold, the threshold is changed (step 5). After changing the threshold value in step 5, the process returns to step 2 again. Step 5, Step 2, Step 3. Until it can be judged as an appropriate threshold value in Step 4. Step 4 is repeated.

  In step 1, a predetermined number of gradations is set as a threshold value. Since it is determined whether or not the threshold value is appropriate in step 4, there is no problem even if any value is set as the threshold value in step 1. For this reason, the setting method at the time of setting this threshold value is not specifically limited, For example, the approximate value which will become a threshold value by visual observation etc. can be set as a threshold value.

  In step 2, one slice image selected from the threshold set in step 1 is binarized. For example, the slice image is converted into a binarized slice image (binarized image) by a method similar to an image conversion step described later. The binarized image is shown in FIG. FIG. 3B shows the image displayed in black and white reversed.

  In step 3, it is calculated how much the resin portion and the filler portion are included in the binarized slice image. For example, the content ratio of the resin portion and the content ratio of the filler portion can be calculated from the binarized slice image by a method of calculating the area ratio between the resin portion and the filler portion. Although the content ratio can be expressed by a quantitative ratio or the like, in the present invention, it is preferable to calculate the content ratio by a calculation method using such an area ratio.

  In step 4, the content ratio of the resin part or filler part obtained in step 3 is compared with the ratio of filler (predetermined ratio) or resin in the actual resin molded product, and set in step 1 It is determined whether or not the threshold value is appropriate. “Ratio of filler in resin molded product (predetermined ratio) or ratio of resin” is a ratio determined from the amount, a ratio determined from the volume, etc., and what ratio is particularly limited Although it is not, it is preferable to express the ratio of each component using a volume ratio. If the ratio between the volume of the resin part in the resin molded product and the volume of the filler part is not known in advance, the specific gravity of the resin molded product is measured and calculated by calculating from the specific gravity of the resin and the specific gravity of the filler. be able to.

  In particular, in this step, the ratio of the resin part or the filler part in step 3 above is expressed as an area ratio, and the ratio of the resin part or the filler part in the actual resin molded product is expressed as a volume ratio. Are preferably compared. Compare the ratio of the occupied volume of the resin part and the occupied volume of the filler part in the actual resin molded product with the ratio of the occupied area of the resin part and the occupied part of the filler part in the binarized slice image By doing so, a more appropriate threshold value can be obtained.

  Furthermore, the threshold value set in Step 1 is an appropriate value based on a comparison between the content ratio of each component obtained in Step 3 and the ratio (predetermined ratio) of the filler in the resin molded product or the ratio of the resin portion. Determine whether or not. One of the features of the present invention is that “a threshold value for discriminating between a filler and a resin portion is determined so that the ratio of the filler in the slice image becomes a predetermined ratio of the filler contained in the actual resin molded product. Step 4 corresponds to the step of determining this specific threshold value. That is, the content ratio of the resin part or the filler part calculated from the binarized slice image is compared with the ratio of the filler in the actual resin molded product (predetermined ratio) or the ratio of the resin, and the threshold is appropriate. It is one of the features of the present invention to determine whether or not. For example, if the ratio of each component is expressed in area ratio in step 3 and the ratio of filler in the resin molded product (predetermined ratio) and the ratio of resin are expressed in volume ratio, the area ratio and volume ratio are as follows. If they match, the threshold value can be determined to be an appropriate value. Even if the area ratio and the volume ratio are different, if the difference can be determined to be a slight difference, the threshold set in step 1 can be determined to be appropriate. How much difference corresponds to “slight difference” depends on the type of resin molded product, etc., but if the difference is within 0.5% when comparing the proportion of resin part or filler part It can be judged as an appropriate threshold value.

  If it is determined that the threshold value set in step 1 is an appropriate threshold value, the threshold value is set as an image density threshold value, and the process proceeds to the next image conversion step. On the other hand, if the threshold value set in step 1 is not determined to be an appropriate threshold value, the process proceeds to step 5 to change the threshold value.

  In step 5, when it is determined that the threshold set in step 4 is not appropriate, the threshold is changed. If it is determined in step 4 that the resin portion is too large to be an appropriate threshold value, the threshold value is changed so that the resin portion is reduced, and in step 4, the filler portion is excessively appropriate. If not, the threshold value is changed so that the filler portion is reduced. Usually, the resin portion is white (bright color) because it easily transmits X-rays compared to the filler portion. Therefore, if the resin portion becomes too much at the threshold value set in step 1, the number of dark gradations is changed to become the threshold value. On the other hand, if the threshold value set in step 1 is too large in the filler portion, the number of brighter gradations is changed to become the threshold value.

  The threshold value is changed, and Steps 2 to 4 are further performed using the threshold value. If the changed threshold value is determined to be an appropriate threshold value in step 4, the process proceeds to the next image conversion step using the changed threshold value as the image density threshold value. On the other hand, if the threshold value is not determined to be appropriate, the threshold value is further changed, and step 5 (threshold value change) and step 2 (slice image are binarized using the changed threshold value) until an appropriate threshold value is obtained. (Converted into an image), step 3 (calculating the ratio of each component from the binarized slice image), and step 4 (determining whether the threshold is appropriate) are repeated.

  In this step, for example, the image density threshold is determined as described above. A method for determining an image density threshold using a plurality of slice images will be described below.

  An image density threshold is determined for each of the plurality of slice images. For example, the image density threshold value can be determined for each slice image by the method of Step 1 to Step 5 as shown in FIG. In this way, it is possible to determine an image density threshold value for each of a plurality of slice images, and to determine a more appropriate image density threshold value by determining an image density threshold value to be used for the next image conversion step from the value of these image density threshold values. it can.

  The method for determining the image density threshold used for the image conversion process from the image density threshold for each of the plurality of slice images is not particularly limited, but a value obtained by averaging all the obtained image density thresholds (average value of the image density thresholds) is used. A method of setting an image density threshold used in the image conversion step is preferable. This is because an appropriate image density threshold can be easily determined.

  As described above, the slice image includes a resin portion, a filler portion, and a noise portion. One of the major features of the orientation state prediction method of the present invention is that a threshold value for separating the resin part and the filler part including the noise part included in the slice image is used, and the ratio of the filler in the slice image is actually resin molding. By determining the image density threshold for discriminating between the filler part and the resin part so that the ratio of the filler contained in the product (predetermined ratio) is reached, the filler content in the resin molded product can be appropriately determined. This is the point that the orientation state can be predicted.

[Image conversion process]
The image conversion step is a step of converting the slice image into a binary image based on the image density threshold and the pixel density of each pixel. A method for converting the slice image into a binarized image is not particularly limited, and examples include a method of displaying a portion above the threshold obtained in the threshold determination step in white and a portion below the threshold in black.

  More specifically, the image density at each pixel determined in the image density determination step is compared with the image density threshold value, and when each pixel has an image density greater than or equal to the image density threshold value, white is selected. FIG. 3B shows a binarized image displaying black when the image density of the pixel is less than the image density threshold. In the binarized image shown in FIG. 3B, the resin portion is expressed in white and the filler portion is expressed in black.

[Reconstruction process]
The reconstruction step is a step of reconstructing a three-dimensional model of a slice image acquisition portion of a resin molded product by sequentially laminating binarized images obtained based on a plurality of slice images in the image conversion step.

  As described in the slice image acquisition step, slice images are acquired at predetermined intervals in a predetermined direction. Therefore, the slice image has a predetermined thickness. Although the resin part and the filler part on the slice image surface can be predicted from the above binarized image, the resin part or the filler part at each position in the thickness direction of the slice image can be reconstructed. It is necessary to predict the filler part. The prediction method at each position in the thickness direction is not particularly limited. For example, in each slice image, when the resin part is predicted on the surface, all the thickness directions are predicted as the resin part, and the filler part is predicted on the surface. In this case, a method of predicting all the thickness directions as filler portions is preferable. In particular, by combining this prediction method and setting the interval between slice images to 20 μm or less as described in the slice image acquisition step, a more accurate three-dimensional model can be easily obtained.

  An example of the three-dimensional model 3 obtained from the orientation state prediction method of the present invention is shown in FIG. The black part in FIG. 5 represents the filler 31, and the white part represents the resin part 32. By using the three-dimensional model obtained from the orientation state prediction method of the present invention for the deformation behavior analysis method described later, accurate deformation behavior analysis becomes possible.

<Deformation behavior analysis method>
The deformation behavior analysis method of the present invention is a deformation behavior analysis performed based on the three-dimensional model obtained by the orientation state prediction method. “Deformation behavior analysis” includes deformation behavior of a resin molded product when a force is applied to the resin molded product, stress generated in the resin molded product, and the like.

  Even if there is only a part of the three-dimensional model in the resin molded product, if the physical properties such as mechanical strength are uniform inside the resin molded product, the entire resin molded product can be repeated in part. By assuming that there is, deformation behavior analysis can be performed. In addition, when a part that is easily deformed, such as a welded part inside a resin molded product, is known in advance, an appropriate deformation behavior analysis is performed by creating a three-dimensional model of the part and analyzing the deformation behavior of that part. It can be performed.

  The deformation behavior analysis method is not particularly limited, and a conventionally known analysis method can be used. The deformation behavior analysis method of the present invention is preferably a deformation behavior analysis method using a finite element method. Hereinafter, an example of deformation behavior analysis using the finite element method will be described.

[Finite element model creation process]
After performing the above-described slice image acquisition process to threshold value determination process, first, the shape definition of the resin molded product to be subjected to deformation behavior analysis is performed. Specifically, a shape is defined by a CAD system or the like, a shape is captured using a CAD interface, or a shape is created by a CAD system. At this time, a rectangular parallelepiped is created in the range imaged by the X-ray CT apparatus.

  After defining the shape, the element is divided by an element division preprocessor to create a model for analysis. At this time, the shape of the element was a cubic element called a voxel element. As the voxel element, a primary element or a secondary element can be selected. It is also possible to divide the voxel element and convert it to a tetrahedral element or the like. In this embodiment, the primary voxel element is used.

[Element model image density determination process]
The image density is determined for each voxel. A slice image is used when determining the image density. The image density is determined from the surface of the slice image. The image density of the slice image is determined as the voxel image density. When the image density of the slice image has light and shade, the method for determining the image density of the voxel is not particularly limited. For example, the image density of the portion having a large occupied area may be determined, or the average of the image density may be taken. .

[Finite element model reconstruction process]
After creating the created voxel element, the positional relationship of each material is projected onto the element to create a binarized finite element model. When projecting, an image density threshold obtained in advance is used.

[Analysis of deformation behavior]
Finally, physical properties and the like are set for each element as boundary conditions. At this time, the physical property value of each component is set according to the purpose and type of analysis such as isotropic property, anisotropy, and nonlinearity. In addition, boundary conditions necessary for structural analysis calculation are given. Then, by calculation based on the finite element method, a desired stress distribution, strain distribution, deformation distribution and the like can be obtained by a conventionally known method.

  Depending on the resolution of the X-ray CT apparatus, the number of element divisions may be large, and depending on the specifications of the finite element method software used, an analysis situation may be impossible or may take a long time. It is necessary to reduce the modeling range according to the specification of the finite element method software, the specification of the computer system to be calculated, and the calculation cost.

  EXAMPLES Hereinafter, although an Example and a comparative example are shown and this invention is demonstrated concretely, this invention is not limited to these Examples.

<Example 1>
[Slice image acquisition process]
The shape of the resin molded product used in Example 1 is a flat plate having ribs on the end surfaces. This model shape 4 is shown in FIG. This resin molded product was produced by injection molding. The shape of the resin molded product and the molding conditions at the time of injection molding are shown below. The gate arrangement in the first embodiment is a two-point pin gate in which a weld is formed.
Shape: Length 100mm, width 50mm, thickness 2mmt, rib thickness 2mmt, rib height 4mmt
Resin: Reinforced polybutylene terephthalate resin containing 30% by mass of glass fiber Mold temperature: 60 ° C.
Resin temperature: 260 ° C
Filling ratio: 17.3 cm ³ / s
Holding pressure: 49 MPa / 5 sec,
Cooling time: 7 sec

Next, the resin molded product obtained by injection molding is cut out in the thickness direction at 3 mm × 10 mm square as shown in FIG. 7B at the position A shown in FIG. The slice image of the resin test piece was acquired by photographing with an X-ray CT apparatus (the test piece includes a weld portion). The resin test piece was photographed on the sample stage of the X-ray CT apparatus. A commercially available X-ray CT apparatus (ELE SCAN mini manufactured by Nippon Steel Elex Co., Ltd.) was used as the imaging apparatus. The shooting conditions are shown below.
Voltage: 32kV
Tube current: 130 mA
Resolution: 6 μm (slice image interval)

[Division process, Image density determination process]
The dividing step is a step of dividing the slice image into pixels having a very small area. In Example 1, the pixels of the slice image are used as they are. For determining the image density of each pixel, commercially available image processing software ImageJ was used.

[Threshold determination step]
The volume fraction of the glass fiber in the examples is 18.1 vol%. This was calculated from the charging ratio of resin and filler at the time of material preparation. The image density threshold value was adjusted so that the ratio of pixels determined to be in the glass range from 17.8% to 18.4% with respect to the number of pixels in the photographed range.

[Image conversion process, reconstruction process]
In order to binarize the image obtained by the X-ray CT apparatus, 1 is set for each pixel constituting the image and the brightness is equal to or higher than the set image density threshold value, and 0 is set for less than that. A three-dimensional model was obtained by assigning a color to each pixel by assigning a color to each pixel, with the image density threshold value (shown by “1”) being black and the image density threshold value (shown by “0”) being white. A three-dimensional model is shown in FIG.

[Deformation behavior analysis method]
As shown in FIG. 8, the finite element model 6 was constructed by dividing the entire resin molded product into 310 elements in the thickness direction, 100 elements in the longitudinal direction of the resin molded product, and 40 elements in the width direction of the resin molded product (see FIG. 8). , After binarization). The shape of the element is a cubic element called a voxel element. The image density of each voxel element was determined from the slice image. At this time, the image density of the voxel was the measured image density in each slice image. When measuring the image density, the above-mentioned commercially available image processing software ImageJ was used.

  Then, material data for one pixel of the binarized image is input to each element. The binarized finite element model obtained by the above operation is shown in FIG. In Example 1, the resin part and the filler part were isotropic, giving an elastic modulus of 2500 MPa and a Poisson's ratio of 0.35 for the resin part, and an elastic modulus of 72000 MPa and a Poisson's ratio of 0.22 for the filler part.

  A detailed calculation method in the above structural analysis is well known in Non-Patent Document 1 and the like, and can be calculated with commercially available software, and thus detailed description thereof is omitted. In Example 1, commercially available software (ANSYS, Inc. ANSYS 11.0) was used. One surface perpendicular to the longitudinal direction of the resin test piece was fixed in the vertical direction, and the other one was subjected to a forced displacement corresponding to 1% strain. The direction perpendicular to the surface was restricted with respect to the surface perpendicular to the width direction of the resin specimen. The analysis was set as linear static analysis. The analysis results (initial elastic modulus (flow direction elastic modulus, flow vertical direction elastic modulus)) are shown in Table 1.

<Example 2>
As Example 2, the same analysis as in Example 1 was performed using X-ray CT image information of a portion not including weld. The shape of the resin molded product used in Example 2 is a flat plate. The shape 7 of the resin molded product is shown in FIG. This resin molded product was produced by injection molding. The shape of the resin molded product and the molding conditions at the time of injection molding are shown below.
Shape: 80 mm long, 80 mm wide, 2 mm thick, 2 mm thick
Resin: Reinforced polybutylene terephthalate resin containing 30% by mass of glass fiber Mold temperature: 60 ° C.
Resin temperature: 260 ° C
Filling rate: 17.3 cm ³ / s
Holding pressure: 39.2 MPa / 5 sec,
Cooling time: 7 sec

  The resin molded product obtained by injection molding is cut out in the thickness direction at 3 mm × 2.5 mm square as shown in FIG. 10B at a position B shown in FIG. The images were taken with X-ray CT in the same manner as in Example 1. The photographing apparatus and the photographing method are the same as those in the first embodiment. Further, the image obtained by the X-ray CT apparatus was binarized as in Example 1. A three-dimensional model of the part photographed with the X-ray CT apparatus was produced. A three-dimensional model 9 is shown in FIG.

  As shown in FIG. 11B, the entire three-dimensional model is divided into 305 elements in the thickness direction, 60 elements in the longitudinal direction of the resin test piece, and 60 elements in the width direction of the resin test piece, thereby constructing a finite element model 10 ( FIG. 11 (b) shows the result after binarization). As the resin part and the filler part, the same material property values as in Example 1 were given as isotropic properties. With respect to the finite element model obtained in this way, one of the surfaces perpendicular to the longitudinal direction of the test piece is fixed in the vertical direction, and the other one corresponds to 1% strain similar to that in Example 1. A forced displacement was given. With respect to a plane perpendicular to the specimen width direction, the plane perpendicular direction was constrained, and analysis was performed using ANSYS 11.0 manufactured by ANSYS, Inc. The analysis was set as linear static analysis. The analysis results (initial elastic modulus (flow direction elastic modulus, flow vertical direction elastic modulus)) are shown in Table 1.

<Comparative Examples 1 and 2>
A comparative example performed by a method using conventional commercially available flow analysis software will be described. A runner and a gate are modeled in the same shape as the first and second embodiments, and the elements are divided. The model was calculated by giving molding conditions as boundary conditions. The molding conditions are the same as in Examples 1 and 2. From the flow velocity distribution, solidification state, etc. of the resin obtained by the flow analysis, the fiber orientation direction and fiber for each element using the Folger-Tucker equation as described in the literature (Moldflow Plastic Insight Ver6.1 help, Moldflow Corp.) Obtain the degree of orientation. Based on the fiber orientation state, a stiffness matrix for each element is calculated using the Tandon-Weng formula as described in the above document. The constraint conditions similar to those of the example were given to the elements, and the conditions were used for the structural analysis. The obtained analysis results (initial elastic modulus (flow direction elastic modulus, flow vertical direction elastic modulus)) are shown in Table 1.

<Reference Example 1>
As an example, in order to verify the effect of the embodiment of the present invention, an actual measurement of physical properties of the weld portion (the test described in Example 1) was attempted. The actual measurement results (initial elastic modulus (flow direction elastic modulus)) are shown in Table 1. The width of the weld portion was 0.6 mm.

<Reference Examples 2 and 3>
In order to compare the accuracy of Example 2 and Comparative Example 2, actual measurement values were obtained. The flat plate obtained by injection molding was cut into a strip shape with the flow plate and the flow vertical direction as the reference and the flat plate center as a reference. Atmospheric temperature was room temperature in accordance with ISO 527-1, 2, and both ends were supported and fixed by a commercially available universal testing machine, and a tensile test was performed to determine initial elastic modulus (flow direction elastic modulus, flow vertical elastic modulus). .

  When the measured value 1 is compared with Example 1 and Comparative Example 1 with respect to the flow direction elastic modulus of the weld portion, Example 1 is closer to the measured value 1 than Comparative Example 1. Regarding the non-weld portion, in Example 2, the elastic modulus is close to the actual measurement value 2, and the accuracy is good. As for Comparative Example 2, the elastic modulus was low overall, and the prediction accuracy was poor.

DESCRIPTION OF SYMBOLS 1 X-ray CT apparatus 11 X-ray irradiation part 12 X-ray detection part 13 Sample stand 14 Rotation drive part 15 Image processing part 2 Resin test piece

Claims (9)

A slice image acquisition step of acquiring a plurality of slice images at predetermined intervals in a predetermined direction for at least a part of a resin molded product formed by molding a resin composition containing a filler at a predetermined ratio;
A division step of dividing the slice image into pixels of a minute area;
An image density determining step for determining an image density of each pixel divided by the dividing step;
One or more slice images are selected from the plurality of slice images, and the filler and the resin portion are discriminated so that the ratio of the filler is the predetermined ratio in the selected slice image. A threshold value determining step for determining an image density threshold value;
An image conversion step of converting the plurality of slice images into a plurality of binarized images based on the image density threshold and the pixel density of each pixel;
Have a, a reconstruction step of said plurality of binary image are sequentially stacked to reconstruct a three-dimensional model of the slice image acquisition portion of the resin molded article,
The threshold determination step includes
A first step of setting a predetermined number of gradations as a threshold;
A second step of binarizing the selected slice image from this threshold;
A third step of calculating how much the resin part and the filler part are respectively included in the binarized slice image;
The content ratio of the resin part or filler part obtained in the third step is compared with the ratio of filler or resin in the actual resin molded product, and the threshold set in the first step is appropriate. A fourth step of determining whether or not,
In the fourth step, when it is determined that the threshold value set in the first step is appropriate, the threshold value set in the first step is determined as an image density threshold value for discriminating between the resin portion and the filler portion. In the fourth step, when it is determined that the threshold value set in the first step is not appropriate, a fifth step of changing the threshold value and returning to the second step;
Including, orientation state prediction method of predicting the orientation of the filler in the resin molded article. The third step is a step of expressing the ratio of the resin part and the filler part as an area ratio,
In the fourth step, the ratio of the filler or the resin in the actual resin molded product is represented by a volume ratio,
In the fourth step, the threshold set in the first step is appropriate when the area ratio and the volume ratio match, or when the difference between the area ratio and the volume ratio is within a predetermined ratio. The orientation state prediction method according to claim 1, wherein the orientation state is predicted. The threshold determination step arbitrarily selects two or more slice images from the slice image, and the filler and the resin are set so that the ratio of the filler is the predetermined ratio in each of the selected slice images. Determining an image density threshold value for discriminating between portions and averaging each image density threshold value to obtain an average value of the image density threshold values;
The orientation according to claim 1 or 2 , wherein the image conversion step is a step of converting the plurality of slice images into a plurality of binarized images based on an average value of the image density threshold values and a pixel density of each pixel. State prediction method. The orientation state prediction method according to claim 1, wherein the slice image is an X-ray CT image. The resin molded product is a resin molded product having a non-fountain flow region,
The orientation according to any one of claims 1 to 4 , wherein the slice image acquisition step is a step of acquiring a plurality of slice images when a portion including the non-fountain flow region is sliced at a predetermined interval in a predetermined direction. State prediction method. The alignment state prediction method according to claim 5 , wherein the non-fountain flow region is a weld portion. A deformation behavior analysis method for analyzing a deformation behavior of a resin molded product using a three-dimensional model obtained from the orientation state prediction method according to claim 1 . A slice image acquisition step of acquiring a plurality of slice images at predetermined intervals in a predetermined direction for at least a part of a resin molded product formed by molding a resin composition containing a filler at a predetermined ratio;
A division step of dividing the slice image into pixels of a minute area;
An image density determining step for determining an image density of each pixel divided by the dividing step;
One or more slice images are selected from the plurality of slice images, and the filler and the resin portion are discriminated so that the ratio of the filler is the predetermined ratio in the selected slice image. A threshold value determining step for determining an image density threshold value;
A finite element model creating step of creating a finite element model of at least a part of the resin molded article;
An element model image density determination step for determining an image density of each of the finite element models;
A finite element model reconstruction step of reconfiguring the finite element model into a finite element model binarized based on the image density threshold and the pixel density of each finite element model ,
The threshold determination step includes
A first step of setting a predetermined number of gradations as a threshold;
A second step of binarizing the selected slice image from this threshold;
A third step of calculating how much the resin part and the filler part are respectively included in the binarized slice image;
The content ratio of the resin part or filler part obtained in the third step is compared with the ratio of filler or resin in the actual resin molded product, and the threshold set in the first step is appropriate. A fourth step of determining whether or not,
In the fourth step, when it is determined that the threshold value set in the first step is appropriate, the threshold value set in the first step is determined as an image density threshold value for discriminating between the resin portion and the filler portion. In the fourth step, when it is determined that the threshold value set in the first step is not appropriate, a fifth step of changing the threshold value and returning to the second step;
including,
A deformation behavior analysis method for analyzing a deformation behavior of a resin molded product using the binarized finite element model. The threshold determination step arbitrarily selects two or more slice images from the slice image, and the filler and the resin are set so that the ratio of the filler is the predetermined ratio in each of the selected slice images. Determining an image density threshold value for discriminating between portions and averaging each image density threshold value to obtain an average value of the image density threshold values;
The finite element model reconstruction process, the image claim the finite element model based mean value and the pixel density of each finite element model of the density threshold is a step of reconstructing the binarizing finite element model 8 Deformation behavior analysis method described in 1.