Mechanical strength prediction method, mechanical strength prediction device, and program

Abstract

The goal is to predict the degree of resin degradation more easily. [Solution] A method for predicting the mechanical strength of a resin, comprising a first prediction step (step S2) of predicting structural information of the resin from color information obtained from the resin, and a second prediction step (step S3) of predicting the mechanical strength of the resin from the structural information of the resin.

Description

Technical Field

[0001] The present invention relates to a method for predicting mechanical strength, a mechanical strength prediction apparatus, and a program.

Background Art

[0002] Conventionally, as a resin composition used as an interior and exterior material for electronic devices including copiers and the like, a thermoplastic resin composition having strength as a resin part and ensuring flame retardancy has been used. Since particularly high strength and flame retardancy are required for exterior materials, materials based on polycarbonate (hereinafter referred to as PC) are often selected. In recent years, from the viewpoints of suppressing global warming and reducing oil consumption, recycling of resin materials has been demanded. For example, an improvement in the recycled material ratio (PCR ratio) in the resin part weight of a copier housing is required, and exterior materials are no exception. [[ID=—16]]Recycled materials of PC are recovered from media such as CDs (Compact Discs) and DVDs (Digital Versatile Discs), gaming machines, bottles, automobiles, building materials, and the like. The recovered PC undergoes a crushing and sorting process, and is provided with functions by compounding as necessary, and then remanufactured. In the sorting process, the target material is immersed in an inorganic solution adjusted to a predetermined specific gravity, and the material within the target specific gravity range is used for remanufacturing, so that the quality is often ensured. Specific gravity sorting is excellent in that it can effectively sort the target resin from resins of different types (specific gravities), or sort alloy materials and single materials, and has high sorting efficiency and is easy to apply in production.

[0003] On the other hand, in recent years, material development utilizing data science such as machine learning has begun to be adopted by research institutions and material development manufacturers. In the formulation design of resin composite materials, a system is also being constructed to predict target physical properties using formulation information and information obtained from measuring instruments as explanatory variables. In Patent Document 1, a method for predicting the degree of deterioration of a certain resin material using factors that promote deterioration as explanatory variables is described. Patent Document 2 describes predicting the molecular weight of an object from the intensity of reflected or transmitted light obtained by irradiating the object with light, and determining whether or not the object is embrittlement based on the predicted molecular weight. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-118496 [Patent Document 2] Japanese Patent Publication No. 2022-73745 [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] By the way, specific gravity sorting is unsuitable for determining the degree of deterioration in resins of the same type where the specific gravity does not change significantly. Furthermore, since Patent Document 1 uses factors that promote deterioration as explanatory variables, it is necessary to recognize the factors that promote deterioration. Patent Document 2 only determines whether or not the material is embrittlement based on the predicted molecular weight, and does not specify the degree of degradation.

[0006] Therefore, the object of the present invention is to predict the degree of resin degradation more easily. [Means for solving the problem]

[0007] To solve the aforementioned problems, the mechanical strength prediction method according to the present invention is: A method for predicting the mechanical strength of a resin using a mechanical strength prediction device, A first prediction step in which structural information of a resin is predicted from color information obtained from the resin, The method includes a second prediction step of predicting the physical properties of the resin from the structural information of the resin.

[0008] Furthermore, the mechanical strength prediction device according to the present invention is A mechanical strength prediction device for predicting the mechanical strength of a resin, A first prediction unit predicts the structural information of a resin from color information obtained from the resin, A second prediction unit predicts the mechanical strength of the resin from the structural information of the resin, It is equipped with.

[0009] Furthermore, the program according to the present invention is A computer for a mechanical strength prediction device that predicts the mechanical strength of resin, A first prediction unit predicts the structural information of a resin from color information obtained from the resin. A second prediction unit predicts the mechanical strength of the resin from the structural information of the resin. To make it function as such. [Effects of the Invention]

[0010] According to the present invention, the degree of resin degradation can be predicted more easily. [Brief explanation of the drawing]

[0011] [Figure 1] This is a block diagram showing the configuration of a mechanical strength prediction device. [Figure 2] This is a flowchart of the machine strength prediction process. [Modes for carrying out the invention]

[0012] Embodiments of the present invention will be described below with reference to the drawings. However, the scope of the present invention is not limited to those described in the following embodiments and drawings.

[0013] <Machine strength prediction device 1> First, the configuration of the mechanical strength prediction device 1 will be explained using Figure 1. The mechanical strength prediction device 1 is an information processing device that predicts the structural information of a resin from color information obtained from the resin, and predicts the mechanical strength of the resin from the structural information of the resin. The resin is, for example, polycarbonate recovered from the market. The resin may also be polycarbonate recovered from used automobiles, gaming machines, media, or bottles. The resin may also be flakes obtained by crushing the recovered polycarbonate.

[0014] Next, the configuration of the mechanical strength prediction device 1 will be described using FIG. 1. As shown in FIG. 1, the mechanical strength prediction device 1 includes a control unit 11, an operation unit 12, a communication unit 13, a storage unit 14, and a display unit 15.

[0015] The control unit 11 is composed of a CPU (Central Processing Unit), a RAM (Random Access Memory), etc. The CPU of the control unit 11 reads out various programs stored in the storage unit 14, expands them in the RAM, executes various processes according to the expanded programs, and controls the operations of each part of the mechanical strength prediction device 1.

[0016] The control unit 11 functions as a first prediction unit that predicts (first prediction) the structural information of the resin from information related to the color obtained from the resin (color information). The information related to the color obtained from the resin (color information) includes any one of the measured value of the color of the resin, CMYK density, lightness L*, chromaticity a*, b*, saturation C*, and hue angle h. The structural information of the resin is the molecular weight of the resin. The first prediction is executed using an analysis model (regression model) such as a machine learning model. The first prediction analysis model is assumed to be trained to output resin structure information as output information when color information is input. Specifically, in the case of polycarbonate recovered from the market, the model is trained on the measured color of the resin obtained from the recovered polycarbonate and the molecular weight of the resin. For example, the resin color can be measured using a fluorescence spectrometer FD-7 (CMYK density, lightness L*, chromaticity a*, b*, saturation C*, hue angle). Also, for example, the molecular weight of the resin can be measured using GPC (Gel Permeation Chromatography).

[0017] The control unit 11 functions as a second prediction unit that predicts the physical properties of the resin (second prediction) from the structural information of the resin. The physical properties of a resin include at least one of the following: impact strength, elastic modulus, and flexural strength. The second prediction is performed using an analytical model (regression model), such as a machine learning model. The analysis model for the second prediction is assumed to be trained to output the mechanical strength of the resin when the structural information of the resin is input.

[0018] The control unit 11 functions as a generation unit that generates formulation information optimized for mechanical strength. Formulation information includes at least one of the following: the ratio of raw materials and the manufacturing conditions of the composite material. Prescription information is generated using analytical models (regression models) such as machine learning models. The analysis model for generating prescription information is assumed to be trained to output optimal prescription information when prescription information and mechanical strength are input. Optimal prescription information is prescription information that satisfies a certain mechanical strength required by the user.

[0019] The operation unit 12 includes a keyboard equipped with cursor keys, number input keys, various function keys, a pointing device such as a mouse, and a touch panel laminated on the surface of the display unit 15. The operation unit 12 is configured to be operable by the operator. The operation unit 12 also outputs various signals to the control unit 11 based on the operations performed by the operator.

[0020] The communication unit 13 is capable of sending and receiving various signals and data with other devices connected via a communication network.

[0021] The memory unit 14 is composed of non-volatile semiconductor memory or a hard disk, and stores various programs executed by the control unit 11, parameters necessary for program execution, and various data. The memory unit 14 stores the analysis models for the first prediction, the second prediction, and the generation of prescription information described above.

[0022] The display unit 15 is composed of a monitor such as an LCD (Liquid Crystal Display) and displays various screens, etc., according to the instructions of the display signals input from the control unit 11.

[0023] <Mechanical strength prediction processing> Next, using Figure 2, we will explain the mechanical strength prediction process performed in the mechanical strength prediction device 1 to predict the mechanical strength of the resin.

[0024] First, the control unit 11 acquires color information (color information) obtained from the resin (step S1). For example, the control unit 11 may acquire color information from the operation unit 12, which is operated by the user. Alternatively, the control unit 11 may acquire color information from a measuring device that measures color information via the communication unit 13.

[0025] Next, the control unit 11 inputs the color information acquired in step S1 into the first prediction analysis model to predict the structural information of the resin (step S2; first prediction step).

[0026] Next, the control unit 11 inputs the resin structure information predicted in step S2 into the analysis model for the second prediction and predicts the mechanical strength of the resin (step S3; second prediction step).

[0027] Next, the control unit 11 outputs the mechanical strength of the resin predicted in step S3 (step S2). For example, the control unit 11 outputs the mechanical strength of the resin to the display unit 15, allowing the user to confirm the predicted mechanical strength of the resin.

[0028] In this way, the degree of resin degradation can be predicted more easily. As mentioned above, specific gravity sorting is highly effective in sorting a target resin from resins of different types (specific gravities) and in sorting alloy materials from single materials. It also has advantages such as high sorting efficiency and ease of application in production. However, it is unsuitable for determining the degree of degradation within the same type of resin where the specific gravity does not change significantly. According to the above method, the degree of degradation within the same resin can be determined by color measurement from a different perspective than specific gravity (weight), and the strength can be predicted.

[0029] <Other> After the mechanical strength prediction process described above, the control unit 11 may perform a generation step to generate optimal formulation information for mechanical strength, and output optimal formulation information, such as the type and amount of additives to adjust various mechanical strengths (impact strength, bending strength, tensile strength, etc.) to the display unit 15 or the like. This makes it easier for users to recreate formulation information for resin composite materials and create resin composite materials that meet the desired mechanical strength.

[0030] <Examples> Using 100 polycarbonate resin flakes derived from water bottles recovered from the market, CMYK density, lightness L*, chromaticity a*, b*, saturation C*, and hue angle h (hereinafter referred to as color measurement data) were measured using a Konica Minolta FD-7 fluorescence densitometer. Furthermore, molecular weight measurements were performed using the same flakes with a Tosoh HLC-8420GPC. The 100 color measurement data points obtained were compressed into a single variable using principal component analysis. Using this variable, a regression equation was created to predict the molecular weight (Mw) of 80 of the 100 flakes. The remaining 20 flakes were predicted using the regression equation, and the model was evaluated using the root mean squared error (RMSE) between the predicted and measured values. Next, strip-shaped test specimens were prepared using polycarbonate resin flakes derived from water bottles recovered from the market, and their molecular weight was measured using the same HLC-8420GPC. Impact tests and bending tests were also performed on the test specimens, and the impact strength, bending strength, and flexural modulus were measured. Using the measured data, a regression equation was created to predict mechanical strength from molecular weight (Mw) data. The mechanical strength of the test specimens was predicted using this regression equation. The model was evaluated using the root mean squared error (Root Mean Squared Error) between the predicted and measured values ​​using the predicted data. The evaluation results are shown in Table I. [Table 1]

[0031] <Effects> As described above, the mechanical strength prediction method is a method for predicting the mechanical strength of a resin, and includes a first prediction step (step S2) of predicting the structural information of the resin from color information (color information) obtained from the resin. The mechanical strength prediction method also includes a second prediction step (step S3) of predicting the mechanical strength of the resin from the structural information of the resin. Therefore, the degree of deterioration can be predicted more easily.

[0032] Furthermore, the color information obtained from the resin includes any of the following: measured color values ​​of the resin, CMYK density, lightness L*, chromaticity a*, b*, saturation C*, or hue angle h. Therefore, even with recovered resin, the degree of degradation can be predicted based on the color of the resin.

[0033] Furthermore, the mechanical strength prediction device 1 is a mechanical strength prediction device that predicts the mechanical strength of a resin, and includes a first prediction unit (control unit 11) that predicts the structural information of the resin from color information (color information) obtained from the resin. Furthermore, the mechanical strength prediction device 1 also includes a second prediction unit (control unit 11) that predicts the mechanical strength of the resin from the structural information of the resin. Therefore, the degree of deterioration can be predicted more easily.

[0034] Furthermore, the program causes the computer of the mechanical strength prediction device 1, which predicts the mechanical strength of the resin, to function as a first prediction unit (control unit 11) that predicts the structural information of the resin from color information (color information) obtained from the resin. Furthermore, the program causes the computer of the mechanical strength prediction device 1 to function as a second prediction unit (control unit 11) that predicts the mechanical strength of the resin from the structural information of the resin. Therefore, the degree of deterioration can be predicted more easily.

[0035] Although the present invention has been described in detail based on embodiments above, it goes without saying that the present invention is not limited to the above embodiments and can be modified as appropriate without departing from the spirit of the invention. For example, the above description disclosed an example in which a hard disk or semiconductor non-volatile memory was used as a computer-readable medium for the program according to the present invention, but the invention is not limited to this example. Portable recording media such as CD-ROMs can be used as other computer-readable media.

[0036] Furthermore, the detailed configuration and operation of each device can be modified as appropriate, without departing from the spirit of the invention. [Explanation of symbols]

[0037] 1. Mechanical strength prediction device 11 Control Unit (First Prediction Unit, Second Prediction Unit, Generation Unit) 12 Control section 13 Communications Department 14 Storage section 15 Display section

Claims

1. A method for predicting the mechanical strength of a resin using a mechanical strength prediction device, A first prediction step in which structural information of a resin is predicted from color information obtained from the resin, A method for predicting mechanical strength, comprising a second prediction step of predicting the physical properties of a resin from the structural information of the resin.

2. The mechanical strength prediction method according to claim 1, wherein the information regarding the color obtained from the resin includes any of the measured color of the resin, CMYK density, lightness L*, chromaticity a*, b*, saturation C*, and hue angle h.

3. The mechanical strength prediction method according to claim 1, wherein the structural information of the resin is the molecular weight of the resin.

4. The mechanical strength prediction method according to claim 1, wherein the physical property value of the resin is at least one of impact strength, elastic modulus, and flexural strength.

5. The method for predicting mechanical strength according to claim 1, wherein the resin is polycarbonate recovered from the market.

6. The mechanical strength prediction method according to claim 1, wherein the resin is PC recovered from used automobiles, amusement machines, media, and bottles.

7. The mechanical strength prediction method according to claim 1, wherein the resin is flakes obtained by crushing recovered polycarbonate.

8. In the first prediction step, a first regression model is used. The mechanical strength prediction method according to claim 1, wherein the first regression model is created by learning the CMYK concentration, lightness L*, chromaticity a*, b*, saturation C*, hue angle h of the resin and the molecular weight of the resin.

9. In the second prediction step, a second regression model is used. The mechanical strength prediction method according to claim 1, wherein the second regression model is created by learning the molecular weight of the resin and the mechanical strength of the resin.

10. The mechanical strength prediction method according to claim 1, comprising a generation step for generating optimal formulation information for mechanical strength.

11. A mechanical strength prediction device for predicting the mechanical strength of a resin, A first prediction unit predicts the structural information of a resin from color information obtained from the resin, A second prediction unit predicts the mechanical strength of the resin from the structural information of the resin, A machine strength prediction device equipped with the following features.

12. A computer for a mechanical strength prediction device that predicts the mechanical strength of resin, A first prediction unit predicts the structural information of a resin from color information obtained from the resin. A second prediction unit predicts the mechanical strength of the resin from the structural information of the resin. A program that makes it function as such.

Images