Apparatus, method, and program

The apparatus and method improve resin molding precision by generating probability distributions and recommending optimal conditions, addressing inaccuracies in existing prediction methods to reduce defective products and enhance manufacturing efficiency.

JP7580449B2Active Publication Date: 2024-11-11ASAHI KASEI KOGYO KABUSHIKI KAISHA
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2022512673
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-03-31
Filing Date
2021-03-31
Publication Date
2024-11-11
Estimated Expiration
2041-03-31

AI Technical Summary

Technical Problem

Existing methods for predicting resin molding conditions suffer from inaccuracies in adjusting molding conditions, leading to an increased occurrence of defective products, as the prediction accuracy varies with the range of molding conditions.

Method used

An apparatus and method that includes a prediction unit to generate probability distributions of predicted resin molded characteristics based on molding factors, displaying these distributions and recommending optimal conditions to achieve target values, using learning models to improve accuracy.

Benefits of technology

Enhances the precision of molding condition adjustments, reducing defective products and saving costs by optimizing resin molding conditions without extensive trial and error, thereby improving manufacturing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007580449000004
    Figure 0007580449000004
  • Figure 0007580449000005
    Figure 0007580449000005
  • Figure 0007580449000006
    Figure 0007580449000006
Patent Text Reader

Abstract

Provided is an apparatus that assists with molding of a resin, the apparatus comprising: a prediction unit that generates a probability distribution for prediction values of to-be-analyzed characteristics of a resin molded article, the prediction values corresponding to the values of a plurality of molding factors in the molding of the resin; and a display processing unit that executes a display process for causing a display device to display the probability distribution for the prediction values of the to-be-analyzed characteristics. When the value of at least one molding factor among the plurality of molding factors in the molding of the resin has changed within a predetermined range, the prediction unit calculates a change in the distribution for the prediction values of the to-be-analyzed characteristics, and the display unit displays the change in the probability distribution for the prediction values of the to-be-analyzed characteristics.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an apparatus, a method, and a program. [Background technology]

[0002] When molding a resin molded product, in order to reduce the occurrence of defective products, the operator of the manufacturing device 2 needs to adjust the molding conditions. Patent Document 1 describes a method of supporting the operator by predicting fluctuations in evaluation items due to molding conditions and displaying them in a graph. Patent Document 1: JP 2006-123172 A

[0003] However, in Patent Document 1, the prediction accuracy of the evaluation items changes depending on the range of molding conditions, which may result in poor accuracy in adjusting the molding conditions.

[0004] In a first aspect of the present invention, there is provided an apparatus for supporting resin molding. The apparatus may include a prediction unit that generates a probability distribution of predicted values ​​of an analysis target characteristic of a resin molded body corresponding to values ​​of a plurality of molding factors of the resin molding. The apparatus may include a display processing unit that executes a display process for displaying the probability distribution of the predicted values ​​of the analysis target characteristic on a display device.

[0005] The prediction unit may calculate a change in probability distribution of the predicted value of the analysis target property when a value of at least one molding factor among a plurality of molding factors of the resin molding is changed within a predetermined range. The display processing unit may execute a display process for displaying the change in probability distribution of the predicted value of the analysis target property on a display device.

[0006] The prediction unit may calculate a change in average value and a change in standard deviation of the predicted value of the analysis target property as indexes showing a change in probability distribution. The display processing unit may execute a display process for displaying the change in average value and the change in standard deviation on a display device.

[0007] The display processing unit may execute a display process for displaying, on a display device, a range of molding factors that have a possibility of satisfying a target value of the analysis target property.

[0008] The apparatus may include an input unit that receives from a user a specification of a value of at least one molding factor among a plurality of molding factors of resin molding. The display processing unit may execute a display process for displaying, on a display device, a probability distribution of a predicted value of the analysis target characteristic at the specified molding factor value together with a probability distribution of the predicted value.

[0009] The prediction unit may have a model that generates a probability distribution of a predicted value of the analysis target characteristic in response to input of data on at least one of a plurality of molding factors of resin molding.

[0010] The prediction section may use a plurality of models to generate respective predicted values ​​of the analysis target characteristic, and calculate an average value and a standard deviation of the generated predicted values.

[0011] The apparatus may include a recommendation unit that generates a recommended condition that is a combination of a plurality of molding factors that has the highest probability of satisfying a target value of the analyzed property. The display processing unit may execute a display process for displaying the recommended condition on the display device together with a change in the probability distribution of the predicted value.

[0012] The prediction unit may include an acquisition unit that acquires learning data including pairs of values ​​of a plurality of molding factors and values ​​of the analysis target property obtained in advance from a resin molding process, and a learning unit that uses the learning data to learn a model that generates a predicted value of the analysis target property from the values ​​of the plurality of molding factors.

[0013] The prediction unit may calculate a probability that at least one molding factor among a plurality of molding factors of the resin molding satisfies a target value of the analyzed property. The display processing unit may execute a display process for displaying, on a display device, the probability distribution of the predicted value and the probability that the analyzed property satisfies the target value.

[0014] The prediction unit may calculate a change in probability that the analysis target property satisfies the target value when a value of at least one of a plurality of molding factors of the resin molding is changed within a predetermined range. The display processing unit may execute a display process for displaying, on a display device, the change in probability that the analysis target property satisfies the target value together with a probability distribution of the predicted value.

[0015] The recommendation unit may generate a plurality of recommended conditions, and each recommended condition may be for a plurality of types of analyzed target characteristics.

[0016] The prediction unit may obtain a probability distribution of predicted values ​​of the corresponding analysis object characteristics from models respectively corresponding to the plurality of analysis object characteristics.

[0017] The recommendation unit may assume that the probability distribution of the predicted values ​​is a Gaussian distribution, and calculate the probability that the analyzed characteristic satisfies a predetermined condition based on the average value and standard deviation of the predicted values. The probability that the analyzed characteristic satisfies the predetermined condition may be the probability that the analyzed characteristic falls below a target value, the probability that the analyzed characteristic exceeds a target value, or the probability that the analyzed characteristic is within a target range.

[0018] The recommendation unit may calculate a joint probability that a combination of a plurality of shaping factors satisfies a predetermined criterion for a plurality of analysis target characteristics, and generate recommended conditions using an optimization algorithm so as to maximize the joint probability.

[0019] The learning unit may add the shaping factor value of the recommended condition generated by the recommendation unit to the learning data as a value of the analysis target characteristic corresponding to the recommended condition that does not satisfy the target value. The recommendation unit may generate a recommended condition different from the recommended condition from a predicted value using the model after learning with the added learning data.

[0020] When a user inputs a specific value of a shaping factor into the input section, the display processing section may display on the display device the probability distribution of the corresponding predicted value superimposed at a position corresponding to the specific value in the display of the change in the probability distribution of the predicted value.

[0021] In a second aspect of the present invention, there is provided a method for assisting resin molding. The method may include a step of generating a probability distribution of predicted values ​​of an analysis target characteristic of a resin molded body corresponding to values ​​of a plurality of molding factors of the resin molding. The method may include a step of executing a display process for displaying the probability distribution of the predicted values ​​of the analysis target characteristic on a display device.

[0022] In the method, the step of generating a probability distribution of a predicted value of the analyzed property may include a step of calculating a change in the probability distribution of a predicted value of the analyzed property of the molded resin when a value of at least one molding factor among a plurality of molding factors of the resin molding is changed within a predetermined range. The step of executing a display process for displaying the probability distribution of the predicted value on a display device may include a step of executing a display process for displaying the change in the probability distribution of the predicted value of the analyzed property on a display device.

[0023] In a third aspect of the present invention, there is provided a program. The program may cause a computer to execute a step of generating a probability distribution of predicted values ​​of an analysis target property of a resin molded body corresponding to values ​​of a plurality of molding factors of the resin molding. The program may cause the computer to execute a step of executing a display process for displaying the probability distribution of the predicted values ​​of the analysis target property on a display device.

[0024] The above summary of the invention does not list all of the necessary features of the present invention. Also, subcombinations of these features may also be inventions. [Brief description of the drawings]

[0025] [Figure 1] 1 shows a system 1 according to the present embodiment. [Diagram 2] 2 shows an example of a screen 200 displayed by the display unit 60 of the present embodiment. [Diagram 3] 2 shows another example of the screen 300 displayed by the display unit 60 of the present embodiment. [Figure 4] 2 shows another example of the screen 400 displayed by the display unit 60 of the present embodiment. [Diagram 5]13 shows a model learning flow in the support device 3 of this embodiment. [Figure 6] 13 shows a display flow of a forming factor in the support device 3 of this embodiment. [Figure 7] 13 shows another example of the screen 500 displayed by the display unit 60 of the present embodiment. [Figure 8] 2 shows another example of the screen 600 displayed by the display unit 60 of the present embodiment. [Figure 9] 13 shows another example of the screen 700 displayed by the display unit 60 of the present embodiment. [Figure 10] 22 illustrates an example computer 2200 in which aspects of the present invention may be embodied, in whole or in part. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention according to the claims. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0027] 1 shows a system 1 according to this embodiment. The system 1 manufactures a resin molded product by adjusting molding conditions for the resin. The system 1 includes a manufacturing device 2 for the resin molded product and a support device 3.

[0028] Examples of the resin composition used in the production of the resin molded product include polyethylene, acrylonitrile, polyamide, etc., or reinforced plastics in which glass fibers or the like are added to resin.

[0029] The manufacturing device 2 is connected to the support device 3 by wire or wirelessly, and manufactures a resin molded product molded into a desired shape by injection molding, in-mold molding, etc. The manufacturing device 2 may evaluate a plurality of analysis target properties of the molded resin (resin molded product) and transmit data indicating a plurality of molding factors and the analysis target properties of the molded resin to the support device 3.

[0030] Here, the multiple molding factors may be factors that affect the quality of the resin molded product, and may include molding conditions set in the manufacturing apparatus 2. The multiple molding factors may be, for example, at least one of mold thickness, mold temperature, cooling time, injection temperature, maximum injection pressure, holding pressure, screw rotation speed, and measurement value. The multiple molding factors may also include at least one of the temperature, humidity, and molded shape of the environment in which the manufacturing apparatus 2 is installed. The values ​​of the multiple molding factors may include discrete values ​​indicating presence / absence (e.g., whether or not a specific material is included in the resin), rank (e.g., rank A, B, or C for the quality or price of the resin), or type of resin, etc.

[0031] The analysis target characteristic may be an evaluation item for determining whether the resin molding is defective or not. The analysis target characteristic may include, for example, at least one of the dimensions, strength, Young's modulus, error from a target dimension of the resin molding, the mode, number, area, or density of defects occurring in the resin molding, and the position where the defects occur.

[0032] The support device 3 is an example of a device that supports resin molding. The support device 3 can support adjustment of molding conditions by displaying predicted values ​​of analysis target properties corresponding to a set of multiple molding factors to a user 4, such as an operator of the manufacturing device 2. The support device 3 includes an acquisition unit 10, a learning unit 20, an input unit 30, a prediction unit 40, a recommendation unit 50, and a display unit 60.

[0033] The acquisition unit 10 is connected to the manufacturing apparatus 2. The acquisition unit 10 acquires learning data including pairs of values ​​of a plurality of molding factors and values ​​of an analysis target property as a result of previously performing resin molding. The acquisition unit 10 may acquire the learning data from at least one of the manufacturing apparatus 2, a website, and an external recording medium. The pairs of values ​​of a plurality of molding factors and values ​​of an analysis target property may be pairs of setting values ​​of a plurality of molding factors when previously performing resin molding in the manufacturing apparatus 2, and values ​​of an analysis target property obtained by evaluating a resin molded product manufactured by the resin molding.

[0034] The learning unit 20 is connected to the acquiring unit 10. The learning unit 20 uses the learning data acquired by the acquiring unit 10 to learn a model that generates a probability distribution of a predicted value of an analysis target characteristic from values ​​of a plurality of shaping factors. The learning unit 20 may generate and update the model using the learning data. The learning unit 20 may store the learning data from the acquiring unit 10 etc. in a database.

[0035] The input unit 30 may be for receiving input from the user 4, and examples thereof include a keyboard, a mouse, or a touch panel. The input unit 30 receives data from the user 4 indicating at least one display condition for displaying on the support device 3. The display condition includes at least one of, for example, at least one value of a plurality of molding factors (for example, a combination of specific values ​​of a plurality of molding factors, a range of at least one of a plurality of molding factors, or a change in the value of at least one of a plurality of molding factors), a designation of at least one analysis target property, and a target value of the analysis target property. The input unit 30 may receive data indicating the display condition from a terminal (such as a personal computer, a tablet, or a smartphone) of the user 4. Furthermore, the input unit 30 may receive an input of molding conditions for performing molding on the manufacturing device 2.

[0036] The prediction unit 40 is connected to the learning unit 20 and the input unit 30. In response to input of data for at least one of the values ​​of the multiple molding factors, the prediction unit 40 receives and holds from the learning unit 20 a model that generates a distribution of predicted values ​​of the analysis target characteristic. The prediction unit 40 generates a distribution of predicted values ​​of the analysis target characteristic of the molded resin corresponding to the values ​​of the multiple molding factors of the resin molding, using the data and model input from at least one of the input unit 30 and the acquisition unit 10. The prediction unit 40 may output the average value or median value and standard deviation of the distribution based on the generated distribution of predicted values. The prediction unit 40 may calculate a change in the probability distribution of the predicted values ​​of the analysis target characteristic when the value of at least one molding factor of the multiple molding factors of the resin molding is changed within a predetermined range (for example, a settable range of the corresponding molding factor). In this case, the prediction unit 40 may calculate a change in the average value or median value and a change in standard deviation of the predicted values ​​of the analysis target characteristic as an index indicating a change in the probability distribution of the predicted values. The prediction unit 40 transmits the probability distribution of the predicted values ​​to the recommendation unit 50. The prediction unit 40 may transmit the learning data acquired by the acquisition unit 10 to the recommendation unit 50.

[0037] The recommendation unit 50 is connected to the prediction unit 40. The recommendation unit 50 generates recommended conditions that are a combination of multiple molding factors that have the highest probability of satisfying the target value of the analysis target property. The recommendation unit 50 may generate recommended conditions in which at least one of the multiple molding factors is changed from a current value (a value currently considered to be optimal). The recommendation unit 50 may generate recommended conditions with at least one molding factor designated by the user 4 via the input unit 30 as the molding factor to be changed. The recommendation unit 50 may transmit the generated recommended conditions to the learning unit 20 as learning data. Furthermore, the recommendation unit 50 may transmit the recommended conditions to the manufacturing apparatus 2 as molding conditions for molding in the manufacturing apparatus 2. When the recommendation unit 50 generates multiple recommended conditions, it may be possible to select one of the generated recommended conditions to be transmitted to the manufacturing apparatus 2. The above selection may be made according to an input received by the input unit 30.

[0038] The display unit 60 is connected to the recommendation unit 50. The display unit 60 has a display processing unit 62 and a display device 64. The display processing unit 62 executes a display process for displaying the probability distribution of the predicted value of the analysis target property generated by the prediction unit 40 on the display device 64. The display processing unit 62 may execute a display process for displaying the recommended conditions on the display device together with the probability distribution of the predicted value. The display processing unit 62 may execute a display process for displaying at least one of the name of the analysis target property, the evaluation conditions, the prediction data, the name of each molding factor, the type name of the resin, the unit, the settable range (for example, the range of the molding factor input via the input unit 30), and the latest set value on the display device 64. The display processing unit 62 may execute a process for generating a display screen, a process for outputting data required for display to the display device 64 wirelessly or wired, and the like. The display device 64 may be a display or the like included in the support device 3, or may be an external display such as a display screen of a terminal included in the user 4 or a display screen of the manufacturing device 2.

[0039] Furthermore, the support device 3 may have a transmission unit 70 that transmits molding conditions to the manufacturing device 2. The transmission unit 70 may receive molding conditions input to the input unit 30 and transmit them to the manufacturing device 2, or may receive recommended conditions generated by the recommendation unit 50 and set them as molding conditions and transmit them to the manufacturing device 2.

[0040] 2 shows an example of a screen 200 displayed by the display unit 60 of this embodiment. On the screen 200, the display unit 60 displays an item name 210 of each molding factor, a slide bar 220 indicating the settable range of the molding factor, a pointer 230 indicating the position of the current value on the slide bar 220, a current value 240 of the molding factor, a list box 250 of the resin type name, and a predicted value box 260 indicating the distribution of predicted values ​​of the analysis target property for the combination of the current values ​​of the molding factors. In FIG. 2, the predicted value box 260 indicates that the average value in the probability distribution of predicted values ​​of the strength of the resin molded product is 100 and the standard deviation is 13.

[0041] While viewing the screen 200, the user 4 uses the mouse (input unit 30) to move the pointer 230 with the mouse cursor to change the value of the molding factor. As a result, the mean and standard deviation of the predicted value distribution in the predicted value box 260 of the analyzed property are changed in real time on the screen by the prediction unit 40 according to the change in the set of values ​​of the molding factors, so that the user 4 can determine molding conditions with high accuracy according to the probability distribution of the predicted values.

[0042] FIG. 3 shows another example of the screen 300 displayed by the display unit 60 of this embodiment. The display unit 60 executes a display process for displaying a change in the probability distribution of the predicted value of the analysis target characteristic on the screen 300 on the display device 64. Specifically, the display processing unit 62 executes a display process for displaying a graph showing a change in the average value and a change in the standard deviation on the display device. As an example, the graph shows the predicted value (strength) of the analysis target characteristic on the vertical axis and the maximum injection pressure (MPa) on the horizontal axis. As an example, the display unit 60 may display a range that may satisfy the target value (strength ≧ 25 in FIG. 3). The display unit 60 may display the screen 300 in a pop-up window that is displayed when the user 4 selects a molding factor (maximum injection pressure in FIG. 3) on the screen 200.

[0043] FIG. 4 shows another example of the screen 400 displayed by the display unit 60 of this embodiment. The display unit 60 may display a plurality of recommended conditions for changing at least one of the multiple molding factors selected by the user 4 from the current value. In the example of FIG. 4, the display unit 60 displays three recommended conditions in the order of first to third place for the drying temperature and drying time. The display unit 60 may display only the molding factors to be changed from the current value as in FIG. 4, or may display values ​​of other molding factors as well. The display unit 60 may display the screen 400 of FIG. 4 on one screen together with at least one of the screen 200 of FIG. 2 and the screen 300 of FIG. 3.

[0044] 5 shows a learning flow of a model in the support device 3 of this embodiment. In step S500, the manufacturing device 2 performs resin molding. The manufacturing device 2 may perform resin molding a plurality of times while changing the molding conditions of the resin molded product.

[0045] In step S510, the acquiring unit 10 acquires learning data from the manufacturing apparatus 2. The acquiring unit 10 may receive a plurality of combinations of values ​​of a plurality of molding factors and values ​​of the analysis target property from the manufacturing apparatus 2. The acquiring unit 10 may also acquire values ​​of a plurality of molding factors and measurement data of the resin molded product (measurement values ​​of dimensions of the molded resin, image data, etc.) from the manufacturing apparatus 2, and the acquiring unit 10 may evaluate the resin molded product to acquire values ​​of the analysis target property.

[0046] In step S520, the learning unit 20 learns a model that generates a probability distribution of a predicted value of an analysis target characteristic from values ​​of a plurality of shaping factors, using the learning data acquired by the acquisition unit 10. The learning unit 20 may receive the recommended conditions from the recommendation unit 50 as learning data, and then learn the model using the learning data. The model is, for example, a Gaussian process method or an ensemble model.

[0047] The learning unit 20 may learn and update the model periodically in response to the acquisition of learning data by the acquisition unit 10 in S510 or in response to an input from the user 4. The learning flow may end when the power of the assistance device 3 is turned off.

[0048] 6 shows a display flow of shaping factors in the support device 3 of this embodiment. In step S600, the input unit 30 may acquire, from the user 4, a designation of a characteristic to be analyzed (e.g., strength, etc.) and at least one designation of a target value of the characteristic to be analyzed. The input unit 30 may also acquire, from the user 4, a designation of a shaping factor whose value is to be changed, values ​​of a plurality of shaping factors (e.g., a combination of current optimal shaping factor values, or changed values, etc.), and a range of a shaping factor (e.g., a settable range, etc.).

[0049] In step S610, the prediction unit 40 uses the model to generate a distribution of predicted values ​​of a specific analysis target characteristic acquired by the input unit 30 from a plurality of shaping factors. The prediction unit 40 may further calculate the average value and standard deviation of the distribution of the generated predicted values. When the model is a Gaussian process method, the learning unit 20 may define a probability distribution from past learning data and output a predicted value probability distribution of the analysis target characteristic for input data such as a combination of shaping factors in a Gaussian distribution. When the model is an ensemble model, the learning unit 20 may use a plurality of models corresponding to the plurality of analysis target characteristics, respectively, to generate predicted values ​​of the analysis target characteristic for input data such as a combination of shaping factors, and output the predicted value probability distribution.

[0050] The prediction unit 40 may generate a plurality of predicted values ​​(or changes) of the analyzed property when a corresponding molding factor among the plurality of molding factors is changed within a predetermined range. The predetermined range to be changed may be, for example, a range input by the user 4 via the input unit 30 for each molding factor, or a settable range of the manufacturing apparatus 2 acquired from the manufacturing apparatus 2 via the acquisition unit 10.

[0051] In step S620, the recommendation unit 50 may generate a recommended condition using the distribution of the predicted value predicted by the prediction unit 40. The recommendation unit 50 may set, as a recommended condition, a combination of forming factors that has a high probability of satisfying the target value of the analysis target characteristic (for example, satisfying a specific range, a range below a threshold, or a range above a threshold, etc.) when one or more of the multiple forming factors are changed within a predetermined range. The recommendation unit 50 may calculate the probability of satisfying the target value of the search analysis target characteristic for each of multiple combinations of multiple forming factors that are candidates for the recommended condition, and determine the recommended condition based on the predetermined condition from the probability. The recommendation unit 50 may receive a selection from the user 4 regarding the forming factor to be changed via the input unit 30. In addition, the recommendation unit 50 may select, as a forming factor to be changed, a forming factor whose change range (range between a minimum value and a maximum value) of the predicted value of the analysis target characteristic exceeds a threshold when the corresponding forming factor is changed within a predetermined range. Furthermore, the recommendation unit 50 may transmit data indicating the recommended conditions to the manufacturing apparatus 2 and control the manufacturing apparatus 2 so as to manufacture the resin molded product according to the recommended conditions.

[0052] In step S620, the display unit 60 displays the probability distribution of the predicted value of the analysis target characteristic and the recommended conditions on the screen. The display unit 60 may display, for example, at least one of the screens shown in FIG. 2, FIG. 3, and FIG. 4 on the same screen or on different screens. The display unit 60 may display a part of the change in the probability distribution of the predicted value from the prediction unit 40. For example, the display processing unit 62 may execute a display process for displaying on the display device 64 the range of molding factors that have a possibility of satisfying the target value of the analysis target characteristic input by the user 4. Here, the analysis target characteristic having a possibility of satisfying the target value may be, for example, a case where the probability distribution of the predicted value of the analysis target characteristic includes the target value or the range of the target value of the analysis target characteristic. In the example of FIG. 3, when the target value of the analysis target characteristic is a strength of 25 or more, the display unit 60 may indicate that the range of the maximum injection pressure in which the average value + standard deviation of the predicted value is 25 or more has a possibility of satisfying the target value, and may indicate that such a range of the maximum injection pressure is x1 or less, between x2 and x3, and x4 or more.

[0053] The display unit 60 may display an optimal combination of values ​​among multiple combinations of multiple molding factors, the most recently acquired current combination of values, or a combination of values ​​input by the user 4 via the input unit 30. The display unit 60 may generate and output screen data and display it on an external display such as the terminal of the user 4 or the manufacturing device 2.

[0054] According to this embodiment, since the probability distribution of the predicted value of the analysis target characteristic can be shown to the user 4, the user 4 can select the value of the molding factor that has a higher probability of satisfying the target value of the analysis target characteristic. This allows the user 4 to easily adjust the molding conditions with high accuracy. In addition, since the optimal molding conditions can be obtained according to the predicted analysis target characteristic without actually performing resin molding with all combinations of molding factors, it is possible to achieve a reduction in manufacturing costs by saving resin material and time. The period from adjusting the molding conditions to starting the production of the product can be shortened, and the operating rate of the manufacturing equipment can be improved.

[0055] Next, an embodiment in which multiple recommended conditions are generated for multiple analysis target characteristics will be described. This embodiment can be executed in combination with at least one step in the flow of Fig. 5 and Fig. 6. In this embodiment, the support device 3 may generate the recommended conditions using an evolutionary algorithm such as a genetic algorithm or simulated annealing.

[0056] The recommendation unit 50 generates a first recommended condition, which is a combination of multiple shaping factors that has the highest probability of satisfying the target value of the analysis target characteristic, in S620 of FIG. 6, and transmits it to the learning unit 20. The learning unit 20 uses the combination of multiple shaping factors of the first recommended condition as learning data to learn a model in S520 of FIG. 5. Next, the prediction unit 40 generates a probability distribution of predicted values ​​using the learned model in S610 of FIG. 6, and transmits it to the recommendation unit 50. Next, the recommendation unit 50 generates a second recommended condition, which is a new combination of multiple shaping factors that has the highest probability of satisfying the target value of the analysis target characteristic, from the probability distribution of predicted values ​​in S620 of FIG. 6. The recommendation unit 50 may further transmit the second recommended condition to the learning unit 20, and generate new recommended conditions in the same manner. Next, a more specific description will be given.

[0057] In this example, the recommended conditions include three types of analysis target properties (for example, y1: number of defects, y2: strength, and y3: Young's modulus), and the recommendation unit 50 generates multiple recommended conditions. First, the prediction unit 40 randomly generates multiple combinations of multiple molding factors that are candidates for recommended conditions within a search range (for example, a settable range) (step 1000). The prediction unit 40 inputs the generated combinations of molding factor values ​​into y1, y2, and y3 models that respectively correspond to the three analysis target properties, and generates predicted average values ​​and predicted standard deviations of the analysis target properties (step 1010). For example, for the i-th combination of molding factors, the predicted average values ​​of the analysis target properties y1, y2, and y3 are calculated as μ i,1 , μ i,2 , and μ i,3 Let the forecast standard deviation be σ i,1 , σ i,2 , and σ i,3 Let us assume that.

[0058] Next, the recommendation unit 50 calculates the probability that the predicted value satisfies the target value of the analysis target characteristic (step 1020). For example, the recommendation unit 50 assumes that the probability distribution of the predicted value is a Gaussian distribution, and calculates a probability P i,1 For the analyzed characteristic y2, the probability that it exceeds b (target value) is P i,2The probability that the analyzed characteristic y3 falls within the range of c to d (target range) is P i,3 is calculated using the following three formulas.

[0059]

number

[0060]

number

[0061]

number

[0062] Next, the recommendation unit 50 calculates the joint probability P i,all =P i,1 P i,2 P i,3 Calculate the joint probability P i,all Candidates for recommended conditions are generated using an optimization algorithm so as to maximize the above-mentioned. The recommendation unit 50 may output a combination of a plurality of molding factors obtained as the recommended conditions as a first recommended condition (step 1030).

[0063] Next, the recommendation unit 50 transmits data indicating the first recommended conditions to the learning unit 20. The learning unit 20 adds the value of the analysis target characteristic for the first recommended conditions as an average value to the learning data database (step 1040). At this time, the learning unit 20 may add the value of the shaping factor of the first recommended conditions to the learning data database as a value of the corresponding analysis target characteristic that does not satisfy the target value. As a result, by using a model trained with the learning data, the recommended conditions obtained will have a combination of values ​​different from those of the first recommended conditions.

[0064] Next, steps 1000-1040 are repeated, and the combination of forming factors obtained in step 1030 at this time is set as the second recommended conditions. Furthermore, steps 1000-1040 can be repeated to obtain third recommended conditions, fourth recommended conditions, and so on.

[0065] According to the present embodiment, it is possible to efficiently generate a plurality of recommended conditions for a plurality of analysis target properties. In addition, it is possible to adjust molding conditions based on the recommended conditions without actually carrying out a large number of resin moldings, thereby achieving reduction in resin molding costs, shortening the time until the product is resin molded, and improving the operation rate of manufacturing equipment.

[0066] In addition, the support device 3 does not need to be connected to the manufacturing device 2, and the acquisition unit 10 may acquire, as learning data, data on resin molding in other manufacturing devices, data recorded on a recording medium, or data on a website.

[0067] Furthermore, the assistance device 3 does not need to include the learning unit 20, and may be provided with a model from an external learning device.

[0068] 9 shows a screen 500 displayed on the display device 64. When the input unit 30 receives from the user 4 a specification of the value of at least one molding factor among a plurality of molding factors for resin molding, the display processing unit 62 may execute a display process for displaying on the display device 64 the probability distribution of the predicted value of the analysis target property at the specified molding factor value together with the probability distribution of the predicted value. The display processing unit 62 may cause the display device 64 to display the probability distribution of the predicted value of the analysis target property at the specified molding factor value in the form of a Gaussian distribution curve used for prediction by the prediction unit 40 in step S610. Note that the vertical axis of the Gaussian distribution curve (the vertical axis direction of the graph) indicates the predicted value of the analysis target property, and the horizontal axis (the horizontal axis direction of the graph) indicates the probability of the predicted value.

[0069] Specifically, in step S620, when the change in the probability distribution (average and standard deviation of the predicted values) of the strength corresponding to the maximum injection pressure is displayed on the display device 64, when the user 4 inputs the maximum injection pressure values ​​of 90 MPa and 265 MPa into the input unit 30, the display processing unit 62 may superimpose and display the probability distribution of the predicted values ​​of the strength at positions on the graph corresponding to the maximum injection pressure values ​​of 90 MPa and 265 MPa.

[0070] Fig. 8 shows a screen 600 displayed on the display device 64. The display processing unit 62 displays the probability distribution of the predicted value of the analysis target characteristic for the specified molding factor value on the display device 64, similar to the screen 500. However, when the prediction unit 40 uses a plurality of models for prediction, a plurality of peaks of high probability may be formed on the probability distribution curve, as shown in Fig. 8. According to the embodiment as shown in Figs. 7 and 8, it is possible to adjust the molding conditions to the appropriate values ​​without actually performing a large number of resin moldings, and it is possible to achieve reduction in resin molding costs, shortening the time until the product is resin molded, improving the operation rate of the manufacturing equipment, and the like.

[0071] 9 shows a screen 700 displayed on the display device 64. The display processing unit 62 may execute a display process for displaying on the display device 64 the probability distribution of the predicted value as well as the probability of satisfying the target value of the analyzed object property. The display processing unit 62 may execute a display process for displaying on the display device 64 the change in the probability of satisfying the target value of the analyzed object property as well as the probability distribution of the predicted value. The prediction unit may calculate the probability of satisfying the target value of the analyzed object property in at least one molding factor among the multiple molding factors of the resin molding, and may calculate the change in the probability of satisfying the target value of the analyzed object property when the value of at least one molding factor among the multiple molding factors of the resin molding is changed within a predetermined range.

[0072] 9 may be similar to the screen of FIG. 3, but additionally includes a change in the probability of meeting the target value. When the user 4 inputs a target value (in FIG. 9, strength ≧25) into the input unit 30, the display processing unit 62 displays a probability P i,2Receive, injection pressure maximum P 0 ~ 300Mpa i,2 The change in probability distribution may be displayed on the display device 64 in a superimposed manner on the change in probability distribution. According to such an embodiment, it is possible to adjust the molding conditions to values ​​with a high probability of being appropriate without actually performing a large number of resin moldings, and it is possible to achieve a reduction in the cost of resin molding, a shortened period until the product is resin molded, an improvement in the operation rate of the manufacturing equipment, etc.

[0073] In addition, the display processing unit 62 may display only the change in the probability of satisfying the target value on the graph, or may display the probability of satisfying the target value at the value of the shaping factor specified by the user 4 at a position corresponding to the change in the probability distribution.

[0074] The support device 3 may be a PC, a server, a mobile terminal, or the like.

[0075] Various embodiments of the present invention may also be described with reference to flow charts and block diagrams, where the blocks may represent (1) stages of a process in which operations are performed or (2) sections of an apparatus responsible for performing the operations. Particular stages and sections may be implemented by dedicated circuitry, programmable circuitry provided with computer readable instructions stored on a computer readable medium, and / or a processor provided with computer readable instructions stored on a computer readable medium. Dedicated circuitry may include digital and / or analog hardware circuitry and may include integrated circuits (ICs) and / or discrete circuits. Programmable circuitry may include reconfigurable hardware circuitry including logical AND, logical OR, logical XOR, logical NAND, logical NOR, and other logical operations, memory elements such as flip-flops, registers, field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like.

[0076] A computer-readable medium may include any tangible device capable of storing instructions that are executed by a suitable device, such that the computer-readable medium having instructions stored thereon comprises an article of manufacture that includes instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, and the like. More specific examples of computer-readable media may include floppy disks, diskettes, hard disks, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), electrically erasable programmable read-only memories (EEPROMs), static random access memories (SRAMs), compact disk read-only memories (CD-ROMs), digital versatile disks (DVDs), Blu-ray (RTM) disks, memory sticks, integrated circuit cards, and the like.

[0077] The computer readable instructions may include either assembler instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0078] Computer readable instructions may be provided to a processor or programmable circuitry of a general purpose computer, special purpose computer, or other programmable data processing apparatus, either locally or over a local area network (LAN), a wide area network (WAN), such as the Internet, etc., to execute the computer readable instructions to create means for performing the operations specified in the flowcharts or block diagrams. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

[0079] 10 illustrates an example of a computer 2200 in which aspects of the present invention may be embodied in whole or in part. Programs installed on the computer 2200 may cause the computer 2200 to function as or perform operations associated with an apparatus or one or more sections of the apparatus according to an embodiment of the present invention and / or to perform a process or steps of the process according to an embodiment of the present invention. Such programs may be executed by the CPU 2212 to cause the computer 2200 to perform specific operations associated with some or all of the blocks of the flowcharts and block diagrams described herein.

[0080] The computer 2200 according to this embodiment includes a CPU 2212, a RAM 2214, a graphic controller 2216, and a display device 2218, which are connected to each other by a host controller 2210. The computer 2200 also includes input / output units such as a communication interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive, which are connected to the host controller 2210 via an input / output controller 2220. The computer also includes legacy input / output units such as a ROM 2230 and a keyboard 2242, which are connected to the input / output controller 2220 via an input / output chip 2240.

[0081] The CPU 2212 operates according to a program stored in the ROM 2230 and the RAM 2214, thereby controlling each unit. The graphic controller 2216 retrieves image data generated by the CPU 2212 into a frame buffer or the like provided in the RAM 2214 or into the graphic controller 2216 itself, and causes the image data to be displayed on the display device 2218.

[0082] The communication interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 in the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0083] The ROM 2230 stores therein a boot program or the like executed by the computer 2200 upon activation, and / or a program that depends on the hardware of the computer 2200. The input / output chip 2240 may also connect various input / output units to the input / output controller 2220 via a parallel port, a serial port, a keyboard port, a mouse port, and the like.

[0084] The programs are provided by a computer-readable medium such as a DVD-ROM 2201 or an IC card. The programs are read from the computer-readable medium, installed in the hard disk drive 2224, the RAM 2214, or the ROM 2230, which are also examples of computer-readable media, and executed by the CPU 2212. Information processing described in these programs is read by the computer 2200, and brings about cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be constructed by realizing information manipulation or processing according to the use of the computer 2200.

[0085] For example, when communication is performed between the computer 2200 and an external device, the CPU 2212 may execute a communication program loaded in the RAM 2214 and instruct the communication interface 2222 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 2212, the communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in the RAM 2214, the hard disk drive 2224, the DVD-ROM 2201, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes reception data received from the network to a reception buffer processing area or the like provided on the recording medium.

[0086] Furthermore, the CPU 2212 may cause all or a necessary portion of a file or database stored in an external recording medium such as the hard disk drive 2224, the DVD-ROM drive 2226 (DVD-ROM 2201), an IC card, etc. to be read into the RAM 2214, and perform various types of processing on the data on the RAM 2214. The CPU 2212 then writes back the processed data to the external recording medium.

[0087] Various types of information, such as various types of programs, data, tables, and databases, may be stored in the recording medium and undergo information processing. The CPU 2212 may perform various types of processing on the data read from the RAM 2214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequence of the program, and write back the results to the RAM 2214. The CPU 2212 may also search for information in a file, database, etc. in the recording medium. For example, when a plurality of entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored in the recording medium, the CPU 2212 may search for an entry that matches a condition, in which an attribute value of the first attribute is specified, from among the plurality of entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0088] The above-described programs or software modules may be stored in a computer-readable medium on the computer 2200 or in the vicinity of the computer 2200. Also, a recording medium such as a hard disk or a RAM provided in a server system connected to a dedicated communication network or the Internet can be used as a computer-readable medium, thereby providing the programs to the computer 2200 via the network.

[0089] Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It is clear to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included in the technical scope of the present invention.

[0090] It should be noted that the order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically indicated as "before," "prior to," etc., and may be realized in any order unless the output of a previous process is used in a later process. Even if the operational flow in the claims, specifications, and drawings is explained using "first," "next," etc. for convenience, it does not mean that it is essential to perform the process in this order. [Explanation of symbols]

[0091] 1 System 2 Manufacturing equipment 3 Support equipment 10 Acquisition Department 20 Learning Department 30 Input section 40 Prediction Department 50 Recommendation section 60 Display 62 Display processing section 64 Display device 70 Transmitter 200 screens 210 Item name 220 Slide bar 230 Pointer 240 Current value 250 List Box 260 Prediction Box 300 screens 400 screens 500 screens 600 screens 700 screens 2200 Computer 2201 DVD-ROM 2210 Host Controller 2212 CPU 2214 RAM 2216 Graphics Controller 2218 Display Device 2220 Input / Output Controller 2222 Communication Interface 2224 Hard Disk Drive 2226 DVD-ROM drive 2230 ROM 2240 Input / Output Chip 2242 Keyboard

Claims

1. An apparatus for supporting resin molding, comprising: a prediction unit that generates a probability distribution of predicted values ​​of analysis target properties of the resin molding body corresponding to values ​​of a plurality of molding factors of the resin molding; a display processing unit that executes a display process for displaying a probability distribution of the predicted value of the analysis object characteristic on a display device, the prediction unit calculates a change in probability distribution of the predicted value of the analysis object characteristic when a value of at least one molding factor among a plurality of molding factors of the resin molding is changed within a predetermined range; The display processing unit executes a display process for displaying, on the display device, a change in the probability distribution of the predicted value of the analysis object characteristic, and executes a display process for displaying, on the display device, a range of the molding factor having a possibility of satisfying a target value of the analysis object characteristic. Device.

2. the prediction unit calculates a change in an average value and a change in a standard deviation of the predicted value of the analysis object characteristic as indexes indicating a change in the probability distribution; The display processing unit executes a display process for displaying the change in the average value and the change in the standard deviation on the display device.

2. The apparatus of claim 1.

3. an input unit for receiving, from a user, a designation of a value of at least one molding factor among the plurality of molding factors of the resin molding; The display processing unit executes a display process for displaying, on the display device, a probability distribution of the predicted value of the analysis target characteristic at the specified value of the forming factor together with the probability distribution of the predicted value.

3. Apparatus according to claim 1 or 2.

4. The prediction unit is a model that generates a probability distribution of the predicted value of the analysis target characteristic in response to input of data on at least one of the plurality of molding factors of the resin molding; 4. Apparatus according to any one of claims 1 to 3.

5. The prediction unit generates the predicted values ​​of the analysis target characteristic using a plurality of the models, and calculates an average value and a standard deviation of the generated predicted values.

5. The apparatus of claim 4.

6. a recommendation unit that generates a recommended condition that is a combination of the plurality of molding factors that is most likely to satisfy a target value of the analysis target property, The display processing unit executes a display process for displaying the recommended conditions on the display device together with a change in the probability distribution of the predicted values.

6. Apparatus according to any one of claims 1 to 5.

7. The prediction unit is an acquisition unit that acquires learning data including pairs of values ​​of the plurality of molding factors and values ​​of the analysis target properties obtained as a result of previously performing resin molding; a learning unit that uses the learning data to learn a model that generates a predicted value of the analysis target property from values ​​of the plurality of molding factors.

7. Apparatus according to any one of claims 1 to 6.

8. the prediction unit calculates a probability that at least one molding factor among a plurality of molding factors of the resin molding satisfies a target value of the analysis target property, The display processing unit executes a display process for displaying, on the display device, a probability distribution of the predicted value and a probability of the analysis target property satisfying a target value.

8. Apparatus according to any one of claims 1 to 7.

9. the prediction unit calculates a change in a probability that the target value of the analysis target property is satisfied when a value of at least one molding factor among a plurality of molding factors of the resin molding is changed within a predetermined range; The display processing unit executes a display process for displaying, on the display device, a probability distribution of the predicted value and a change in the probability of satisfying the target value of the analysis object characteristic.

9. The apparatus of claim 8.

10. 1. A method for assisting resin molding, comprising: generating a probability distribution of predicted values ​​of analysis target properties of the resin molding corresponding to values ​​of a plurality of molding factors of the resin molding; and executing a display process for displaying a probability distribution of the predicted value of the analysis object property on a display device, The step of generating a probability distribution of the predicted values ​​comprises: a step of calculating a change in probability distribution of the predicted value of the analysis target characteristic when a value of at least one of a plurality of molding factors of the resin molding is changed within a predetermined range, The step of executing the display process includes: a display process for displaying on the display device a change in the probability distribution of the predicted value of the analysis object characteristic, and a display process for displaying on the display device a range of the forming factor having a possibility of satisfying a target value of the analysis object characteristic. method.

11. the step of generating a probability distribution of the predicted value of the analysis target characteristic includes a step of calculating a change in the probability distribution of the predicted value of the analysis target characteristic of the resin molded body when a value of at least one molding factor among a plurality of molding factors of the resin molding is changed within a predetermined range; The step of executing the display process includes a step of executing a display process for displaying a change in the probability distribution of the predicted value of the analysis target characteristic on the display device. The method of claim 10.

12. On the computer, A step of generating a probability distribution of predicted values ​​of an analysis target characteristic of a resin molding corresponding to values ​​of a plurality of molding factors of the resin molding; and executing a display process for displaying a probability distribution of the predicted value of the analysis object characteristic on a display device. The step of generating a probability distribution of the predicted values ​​comprises: a step of calculating a change in probability distribution of the predicted value of the analysis target characteristic when a value of at least one of a plurality of molding factors of the resin molding is changed within a predetermined range, The step of executing the display process includes: a display process for displaying on the display device a change in the probability distribution of the predicted value of the analysis object characteristic, and a display process for displaying on the display device a range of the forming factor having a possibility of satisfying a target value of the analysis object characteristic. program.

Citation Information

Patent Citations

  • Preparation method of plant fiber felt / resin-based composite material

    CN109719973A

  • Molding condition searching method engaging expert system

    JP1992035923A

  • Controller device of injection molding machine

    JP2006281662A

  • Support device of injection molding machine

    JP2008110486A

  • Molding support method of injection molder

    JP2017081071A