Learning device, estimation device, estimation system, learning method, estimation method, and program

A machine learning model predicts 28-day strength of sabo soil cement using 7-day data, addressing inefficiencies in conventional testing by shortening the process and reducing costs and delays.

JP2024158012AActive Publication Date: 2024-11-08INVAX CORP

Patent Information

Application Number
JP2023072787
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-26
Publication Date
2024-11-08
Estimated Expiration
2043-04-26

AI Technical Summary

Technical Problem

The conventional sabo soil cement mixing tests require a minimum of 28 days to obtain strength results, necessitating repeated tests until desired quality is achieved, and often involve transporting soil to a facility for steam curing, which is inefficient and costly.

Method used

A machine learning-based approach using a strength estimation model that predicts the 28-day strength of sabo soil cement by inputting 7-day strength data, reducing the need for extensive physical testing and enabling on-site evaluations.

Benefits of technology

This method significantly shortens the testing period from 28 days to 7 days, reduces soil collection and transportation needs, and allows for more efficient formulation determination without additional tests, thereby reducing costs and construction delays.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024158012000001_ABST
    Figure 2024158012000001_ABST
Patent Text Reader

Abstract

To shorten time required for soil cement mixing test.SOLUTION: A learning control unit learns a strength estimation model through machine learning using teacher data indicating a solidification agent amount and water content of each of a plurality of types of soil cement mixing cases, and a first period strength that is a strength after a first period elapses and a second period strength that is a strength after a second period that is longer than the first period elapses, of the soil cement in each of the plurality of types of mixing cases. An estimation unit inputs a solidification agent amount and water content of each of a plurality of types of mixing cases of soil cement whose second period strength is to be estimated and a first period strength of one or more mixing cases, into the strength estimation model, and obtains an output of a second period strength of the soil cement of each of the plurality of types of mixing cases having the input solidification agent amount and water content.SELECTED DRAWING: Figure 12
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a learning device, an estimation device, an estimation system, a learning method, an estimation method, and a program. [Background technology]

[0002] Due to abnormal weather worldwide, landslides are occurring frequently. In recent years, in disaster recovery sites, instead of the conventional concrete sabo dams, sabo dams made of sabo soil cement, which is made by firmly solidifying local soil and sand, are being used.

[0003] This method differs from the soil cement method that has been used in conventional ground improvement work in that it uses mainly soil washed away by debris flows and residual soil generated during construction. With the soil cement erosion control method, it is extremely important to carry out an appropriate mix design, including the type and amount of solidifying agent (cement, etc.) that matches the material properties and the water content setting, so that the required strength is achieved. Mix design tests are conducted to carry out this mix design.

[0004] The outline of a typical mix test for sabo soil cement is as follows. (1) 500kg to 600kg of soil will be collected on-site, and after conducting soil tests according to JIS standards, nine basic mix combination cases will be considered. (2) In the testing laboratory, six specimens are prepared for each of the nine blending cases, for a total of 54 specimens. (3) For each mix case, the 7-day strength of three specimens and the 28-day strength of three specimens are measured, and the results are analyzed to determine the appropriate nominal mix. (4) If the results obtained in (3) do not satisfy the required quality, reconsider the mix case, conduct additional tests from (2), analyze the results, and determine the appropriate mix.

[0005] Meanwhile, there is a technique for determining early whether the strength of sabo soil cement is equal to or greater than a target strength (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent Publication No. 2021-9062 Summary of the Invention [Problem to be solved by the invention]

[0007] As mentioned above, in the mix test, additional tests are repeatedly conducted until test results that satisfy the required quality are obtained. However, one mix test requires at least 28 days to obtain strength, and if there is insufficient soil for testing, soil must be collected from the site again. In addition, in the technology of Patent Document 1, the strength is measured after the test specimen is steam cured at a specified curing temperature for a specified time. Therefore, it was necessary to transport the soil to a facility where the test specimen can be steam cured and then conduct the test.

[0008] The present invention has been made in consideration of the above-mentioned circumstances, and provides a technique that makes it possible to shorten the time required for soil cement mix testing. [Means for solving the problem]

[0009] One aspect of the present invention is a learning device that includes a learning control unit that uses teacher data indicating the amount of solidifying agent and water content of each of multiple types of mixing cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a first period has elapsed, and a second period strength, which is the strength after a second period longer than the first period, to learn a strength estimation model through machine learning, which takes as input the amount of solidifying agent and water content of each of multiple types of mixing cases and the first period strength of the soil cement of one or more of the mixing cases, and outputs the second period strength of the soil cement of each of the multiple types of mixing cases having the input amount of solidifying agent and water content.

[0010] One aspect of the present invention is the above-mentioned learning device, wherein the teacher data further includes information on the type of solidification material, and the learning control unit inputs the type of solidification material, the amount of solidification material and water content of each of multiple types of mixing cases, and the first period strength of the soil cement in one or more of the mixing cases, and learns through machine learning a strength estimation model that outputs the second period strength of the soil cement in each of the multiple types of mixing cases with the input amount of solidification material and water content.

[0011] One aspect of the present invention is an estimation device that includes an estimation unit that inputs the solidification agent amount and water content of each of multiple types of mixing cases that are the subject of second period strength estimation and the first period strength of the soil cement of one or more of the multiple types of mixing cases that are the subject of second period strength estimation into a strength estimation model trained using teacher data indicating the solidification agent amount and water content of each of multiple types of mixing cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a first period has elapsed, and a second period strength, which is the strength after a second period that is longer than the first period has elapsed, and obtains an output of the second period strength of the soil cement of each of the multiple types of mixing cases that have the input solidification agent amount and water content.

[0012] One aspect of the present invention is the above-mentioned estimation device, wherein the teacher data further includes information on the type of solidification material, and the estimation unit inputs into the strength estimation model the type of solidification material, the amount of solidification material and water content of each of the multiple types of mixing cases for which second period strength is to be estimated, and the first period strength of the soil cement in one or more of the multiple types of mixing cases for which second period strength is to be estimated, thereby obtaining an output of the second period strength of the soil cement in each of the multiple types of mixing cases with the input amount of solidification material and water content.

[0013] One aspect of the present invention is an estimation system comprising: a learning control unit that uses teacher data indicating the amount of solidifying agent and water content of each of multiple types of mixing cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a first period, and a second period strength, which is the strength after a second period longer than the first period, to learn a strength estimation model by machine learning, which inputs the amount of solidifying agent and water content of each of multiple types of mixing cases and the first period strength of the soil cement of one or more of the mixing cases, and outputs a second period strength of the soil cement of each of the multiple types of mixing cases having the input solidifying agent amount and water content; and an estimation unit that inputs the amount of solidifying agent and water content of each of the multiple types of mixing cases that are the subject of second period strength estimation, and the first period strength of the soil cement of one or more of the multiple types of mixing cases that are the subject of second period strength estimation, into the strength estimation model, to obtain an output of the second period strength of the soil cement of each of the multiple types of mixing cases having the input solidifying agent amount and water content.

[0014] One aspect of the present invention is a learning method comprising a learning step of using teacher data indicating the amount of solidifying agent and water content of each of multiple types of mixing cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a first period has elapsed, and a second period strength, which is the strength after a second period longer than the first period, to learn a strength estimation model by machine learning, which inputs the amount of solidifying agent and water content of each of multiple types of mixing cases and the first period strength of the soil cement of one or more of the mixing cases, and outputs a second period strength of the soil cement of each of the multiple types of mixing cases having the input amount of solidifying agent and water content.

[0015] One aspect of the present invention is an estimation method having an estimation step of inputting the solidification agent amount and water content of each of multiple types of mixing cases for which a second period strength is to be estimated and the first period strength of the soil cement of one or more of the multiple types of mixing cases for which a second period strength is to be estimated into a strength estimation model trained using teacher data indicating the solidification agent amount and water content of each of multiple types of mixing cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a first period has elapsed, and a second period strength, which is the strength after a second period longer than the first period has elapsed, thereby obtaining an output of the second period strength of the soil cement of each of the multiple types of mixing cases for which the solidification agent amount and water content have been input.

[0016] One aspect of the present invention is a program for causing a computer to function as the above-mentioned learning device.

[0017] One aspect of the present invention is a program for causing a computer to function as the above-described estimation device. Effect of the Invention

[0018] The present invention makes it possible to shorten the time required for soil cement mix testing. [Brief description of the drawings]

[0019] [Figure 1] A diagram showing examples of blending cases subject to blending testing. [Diagram 2] FIG. 1 is a flow diagram showing the procedure for carrying out a blending test. [Diagram 3] FIG. 1 is a diagram showing how a blending test is carried out. [Figure 4] FIG. 1 is a diagram showing how a blending test is carried out. [Diagram 5] FIG. 1 is a diagram showing how a blending test is carried out. [Figure 6] FIG. 1 is a diagram showing how a blending test is carried out. [Figure 7] FIG. 1 is a diagram showing how a blending test is carried out. [Figure 8]FIG. 1 is a diagram showing how a blending test is carried out. [Figure 9] FIG. 1 is a diagram showing how a blending test is carried out. [Figure 10] FIG. 13 is a diagram showing an example of strength distribution due to differences in water content between soil and sabo soil cement. [Figure 11] FIG. 1 is a diagram showing an example of the relationship between the water content of sabo soil cement and its 28-day strength. [Figure 12] 1 is a schematic block diagram illustrating a system configuration of an estimation system according to an embodiment. [Figure 13] 2 is a schematic block diagram showing a specific example of a functional configuration of a terminal device according to an embodiment. FIG. [Figure 14] FIG. 4 is a diagram illustrating an example of teacher data according to the embodiment. [Figure 15] 2 is a schematic block diagram showing a specific example of the functional configuration of a learning device according to an embodiment. FIG. [Figure 16] 11 is a flowchart illustrating a specific example of processing performed by the learning device according to the embodiment. [Figure 17] 1 is a schematic block diagram showing a specific example of a functional configuration of an estimation device according to an embodiment. [Figure 18] 10 is a flowchart illustrating a specific example of a process performed by the estimation device according to the embodiment. [Figure 19] FIG. 11 is a diagram illustrating an example of a screen displayed on a terminal device according to the embodiment. [Figure 20] FIG. 2 is a diagram illustrating an outline of a hardware configuration example of an information processing device applied to the present embodiment. [Figure 21] FIG. 11 is a schematic block diagram showing a modified example of the functional configuration of the estimation device according to the embodiment. [Figure 22] FIG. 11 is a schematic block diagram showing a modified example of the functional configuration of the estimation device according to the embodiment. [Figure 23] FIG. 11 is a diagram showing an estimation result using the estimation system of the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. First, the implementation procedure of the mixing test of the sand control soil cement will be described. The sand control soil cement is an example of soil cement. Here, a case where nine types of mixing cases are considered after conducting a soil quality test according to JIS standards on the earth and sand to be tested will be described as an example.

[0021] Figure 1 shows an example of the mixing cases for which the mixing test is to be conducted. In the nine types of mixing cases shown in Figure 1, the same type of solidifying material is used. Hereinafter, the case where the solidifying material is cement will be described as an example. First, three cement cases, namely CASE1, CASE2, and CASE3, are set. The unit cement amounts of CASE1, CASE2, and CASE3 are C1, C2, and C3 respectively. Then, three types of water contents, Wn1, Wn2, and Wn3, are set for each cement case CASEn (n = 1, 2, 3), and they are respectively designated as CASEn-1, CASEn-2, and CASEn-3. As a result, a total of nine mixing cases, CASE1-1 to 3-3, are set. The unit cement amount satisfies C1 < C2 < C3, and the water contents satisfy W11 < W12 < W13, W21 < W22 < W23, and W31 < W32 < W33. It is often the case that C2 - C1 = C3 - C2, but it is not limited to this. Also, it is often the case that W12 - W11 = W13 - W12 = W22 - W21 = W23 - W22 = W32 - W31 = W33 - W32, but it is not limited to this. In the mixing test, generally, the 7-day strength and 28-day strength of the sand control soil cement for each of these nine mixing cases are measured.

[0022] Figure 2 is a flowchart showing the procedure of the mixing test. Conventionally, for each of the nine mixing cases, the procedures from step S1 to step S9 are carried out. Also, Figures 3 to 9 are diagrams showing the implementation states of the respective procedures shown in Figure 2.

[0023] First, the test performer confirms the mixing amount of the materials for the sand control soil cement for which the mixing test is to be conducted (step S1, Figure 3). The materials are the sample, which is the earth and sand to be tested, cement, and added water. The mixing amounts of these materials are determined based on the type of cement, the unit cement amount, and the water content of each mixing case shown in Figure 1.

[0024] Next, the tester puts the sample and cement into the mixer for each mix case and performs dry mixing for the specified time (step S2, Figure 4). Next, the tester puts the added water into the mixer and performs main mixing for the specified time (step S3, Figure 5). After mixing in the main mix, the tester checks the mixed state visually and by touch (step S4, Figure 6). The tester measures the moisture content of the mixed sample to check the consistency between the test moisture content and the moisture content after mixing (step S5, Figures 7(a) and 7(b)). Next, the tester measures the mixed sample using a slump cone (step S6, Figure 8).

[0025] After that, the tester prepares three specimens for a 7-day compressive strength test and three specimens for a 28-day compressive strength test for each of the nine mix cases (Step S7). In preparing these specimens, the mixed sample is placed in a mold and then compacted in the mold by tamping with a ram and compacting with a vibrator (Fig. 9(a)). The sample is placed and compacted in several steps. After the specimens are prepared (Fig. 9(b)), the tops are capped with mortar or the like as necessary. The strength of sabo soil cement is affected by the unit cement amount, water content, curing temperature, moisture management, etc., so these factors are kept constant in each mix test.

[0026] The tester measures the compressive strength of the three 7-day compressive strength test specimens for each mix case on the seventh day after the specimens are made to obtain the 7-day strength (step S8). Furthermore, the tester measures the compressive strength of the three 28-day compressive strength test specimens for each mix case on the 28th day after the specimens are made to obtain the 28-day strength (step S9).

[0027] The purpose of the mix tests is to confirm the 28-day strength for each mix case. However, because the purpose of sabo soil cement is to harden through cement hydration (chemical reaction), and the increase in strength from 7 days to 28 days is important data when judging the recommended mix (a mix that satisfies the required quality), the 7-day strength of the specimens is also measured. Note that for each type of mix case, three specimens are prepared for the 7-day compressive strength test and three for the 28-day compressive strength test; however, this is to make it possible to detect any abnormal values ​​obtained, and the number is not limited to three.

[0028] FIG. 10 is a diagram showing an example of strength distribution according to the difference in the water content of soil and sabo soil cement. Curve A0 shows the relationship between the water content of soil and compaction strength, and curves A1, A2, and A3 are quadratic curves obtained from the 28-day strength against the water content of sabo soil cement with the same cement type and unit cement amount. Curve A0 is obtained, for example, by material testing. w1, w2, and w3 are the water contents of sabo soil cements CASEn-1, CASEn-2, and CASEn-3, respectively, and C1, C2, and C3 are the 28-day strengths of CASEn-1, CASEn-2, and CASEn-3, respectively. In addition, P is the peak strength, and B is the minimum controlled strength. The minimum controlled strength is the strength at which it is guaranteed that the material has hardened due to the hydration reaction.

[0029] Figure 10(a) shows the measurement results of 28-day strength when the strength peak P was captured in CASEn-1 to n-3. In this case, peak P is on curve A1. Figure 10(b) shows the measurement results when the amount of water added exceeds the strength peak P in CASEn-1 to n-3. Curve A2 decreases as the water content increases. Figure 10(c) shows the measurement results when the amount of water added falls below the strength peak in CASEn-1 to n-3. Curve A3 decreases as the water content increases. If peak P cannot be captured from the 28-day strength results as in Figures 10(b) and 10(c), the mix may be reconsidered and mix tests may be conducted so that peak P can be captured.

[0030] FIG. 11 is a diagram showing an example of the relationship between the water content and 28-day strength of the erosion control soil cement obtained from each of the cement cases, CASE 1, 2, and 3. Curve A0 shows the relationship between the water content of the soil and compaction strength, curves D1, D2, and D3 show the relationship between the water content and 28-day strength of the erosion control soil cement obtained from CASE 1, 2, and 3, respectively, and α1, α2, and α3 show the peak strength and water content of the erosion control soil cement from CASE 1, 2, and 3, respectively. FIG. 11(a) shows a case where the peak strengths of all three cement cases exceed the minimum control strength, and FIG. 11(b) shows a case where the peak strengths of two cement cases exceed the minimum control strength. In these cases, two or more peak strengths (for example, 2 N / mm 2 Based on these peak strengths, it is possible to determine the recommended mix proportions. On the other hand, Figure 11(c) shows a case where the peak strength of one cement case exceeds the minimum control strength, and Figure 11(d) shows a case where the peak strength of none of the three cement cases exceeds the minimum control strength. In these cases, mix cases may be reconsidered and mix tests may be conducted so that two or more peak strengths can be obtained. In the past, repeated mix tests would increase the test period and could also cause problems in terms of increased costs.

[0031] Next, the estimation system of this embodiment will be described. FIG. 12 is a schematic block diagram showing a system configuration of an estimation system 100 according to one embodiment of the present invention. The estimation system 100 includes a terminal device 10, a learning device 20, and an estimation device 30. The terminal device 10 and the estimation device 30 are communicatively connected via a network 70. The learning device 20 and the estimation device 30 may be communicatively connected via the network 70. The network 70 may be a network using wireless communication, or may be a network using wired communication. The network 70 may be configured using, for example, the Internet, or may be configured using a local area network (LAN). The network 70 may be configured by combining a plurality of networks.

[0032] The learning device 20 uses blending test data showing the results of blending tests conducted in the past as training data, and machine-learns a strength estimation model according to the procedure shown in Fig. 2. The strength estimation model receives as input the type of cement, the unit cement amount (solidifying agent amount) and water content for each of nine blending cases, and the 7-day strength of the erosion control soil cement for one or more blending cases, and outputs the 28-day strength of the erosion control soil cement for each of the nine blending cases with the input unit cement and water content.

[0033] When a mix test is conducted on new soil or sand, the tester considers nine mix cases for the soil or sand, and measures the 7-day strength of the erosion control soil cement in at least one of those mix cases according to the procedure shown in Fig. 2. The estimation device 30 inputs the type of cement, the unit cement amount and water content of each of the nine mix cases considered, and information on the 7-day strength of one to nine of the considered mix cases (hereinafter referred to as "input information") into the strength estimation model after learning by the learning device 20 (hereinafter also referred to as "trained model"), and obtains as output an estimated value of the 28-day strength of the erosion control soil cement in each of the nine mix cases with the input unit cement and water content.

[0034] 13 is a schematic block diagram showing a specific example of the functional configuration of the terminal device 10. The terminal device 10 is configured using information devices such as a smartphone, a tablet, a personal computer, a dedicated device, etc. The terminal device 10 includes a communication unit 11, an input unit 12, an output unit 13, a storage unit 14, and a control unit 15.

[0035] The communication unit 11 is a communication device. The communication unit 11 may be configured as, for example, a network interface. The communication unit 11 performs data communication with other devices via the network 70 in response to the control of the control unit 15. The communication unit 11 may be a device that performs wireless communication or a device that performs wired communication.

[0036] The input unit 12 is configured using existing input devices such as a keyboard, a pointing device (mouse, tablet, etc.), buttons, a touch panel, etc. The input unit 12 is operated by a user when inputting a user's instruction to the terminal device 10. The input unit 12 may be an interface for connecting the input device to the terminal device 10. In this case, the input unit 12 inputs an input signal generated in the input device in response to a user's input to the terminal device 10.

[0037] The output unit 13 outputs information in a form that can be recognized by the user. The output unit 13 may be, for example, an image display device such as a liquid crystal display or an organic EL (Electro Luminescence) display. The output unit 13 may be an interface for connecting the image display device to the terminal device 10. In this case, the output unit 13 generates a video signal for displaying image data, and outputs the video signal to the image display device connected to the output unit 13. The output unit 13 may be a device for outputting sound, such as a speaker. The output unit 13 may be an interface for connecting an audio output device, such as a speaker or a headphone, to the terminal device 10. In this case, the output unit 13 generates an audio signal for reproducing audio data, and outputs the audio signal to the audio output device connected to the output unit 13. The output unit 13 may be configured as a touch panel integrated with the input unit 12.

[0038] The storage unit 14 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. The storage unit 14 stores data used by the control unit 15. The storage unit 14 stores data required when the control unit 15 performs processing.

[0039] The control unit 15 is configured using a processor such as a CPU (Central Processing Unit) and a memory (main storage device). The control unit 15 functions by the processor executing a program. All or part of the functions of the control unit 15 may be realized using hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array). The above program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a semiconductor storage device (e.g., an SSD: Solid State Drive), or a storage device such as a hard disk or a semiconductor storage device built into a computer system. The above program may be transmitted via an electric communication line.

[0040] The control unit 15 may execute, for example, an application installed in its own device (terminal device 10). A specific example of such an application is an application provided to the terminal device 10 as a dedicated application for the estimation system 100. Another specific example of such an application is a WEB (web) browser application. Such an application may be installed in the terminal device 10 in advance, or may be downloaded each time an estimation process is executed. For example, when implemented as a WEB browser application, the terminal device 10 may download and execute the application from a device (for example, the WEB server itself or another server) designated by the WEB server in response to the terminal device 10 connecting to the specific WEB server. The control unit 15 operates according to the program of the application being executed.

[0041] The control unit 15 controls the terminal device 10 in response to a user's operation or information received from the estimation device 30. For example, the control unit 15 transmits information input by the user operating the input unit 12 to the estimation device 30 by using the communication unit 11. For example, when the information transmitted from the estimation device 30 is received by the communication unit 11 via the network 70, the control unit 15 generates screen data based on the received information and causes the output unit 13 to display the screen data. Such screen data includes images and characters indicating the information transmitted from the estimation device 30. For example, when the information transmitted from the estimation device 30 is received by the communication unit 11 via the network 70, the control unit 15 may generate voice data based on the received information and cause the output unit 13 to output the voice data.

[0042] 14 is a diagram showing an example of training data. The training data includes the type of cement, unit cement amount, water content, 7-day strength of three test specimens, and 28-day strength of three test specimens for each of the past training tests for the respective blending cases 1-1 to 3-3.

[0043] 15 is a schematic block diagram showing a specific example of the functional configuration of learning device 20. Learning device 20 is configured using an information processing device such as a personal computer or a server device. Learning device 20 includes a communication unit 21, a storage unit 22, and a control unit 23.

[0044] The communication unit 21 is a communication device. The communication unit 21 may be configured as, for example, a network interface. The communication unit 21 performs data communication with other devices via the network 70 in accordance with the control of the control unit 23. The communication unit 21 may be a device that performs wireless communication or a device that performs wired communication.

[0045] The storage unit 22 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. The storage unit 22 stores data used by the control unit 23. The storage unit 22 may function as, for example, a teacher data storage unit 221, a preprocessed teacher data storage unit 222, and a trained model storage unit 223.

[0046] The teacher data storage unit 221 stores teacher data used in the learning process executed by the learning device 20. The teacher data stored in the teacher data storage unit 221 is combination test data indicating the results of each past combination test as shown in FIG.

[0047] The preprocessed teacher data storage unit 222 stores preprocessed teacher data. The preprocessed teacher data stores information obtained by the preprocessing control unit 232 executing preprocessing on the teacher data. The preprocessing is a process of converting the teacher data into a data format for inputting the teacher data into the intensity estimation model.

[0048] The trained model storage unit 223 stores a trained model obtained by a learning process using the teacher data stored in the teacher data storage unit 221 and the preprocessed teacher data stored in the preprocessed teacher data storage unit 222.

[0049] The control unit 23 is configured using a processor such as a CPU and a memory. The control unit 23 functions as an information control unit 231, a pre-processing control unit 232, and a learning control unit 233 by the processor executing a program. All or part of the functions of the control unit 23 may be realized using hardware such as an ASIC, a PLD, or an FPGA. The above program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a semiconductor storage device (e.g., SSD), or a storage device such as a hard disk or a semiconductor storage device built into a computer system. The above program may be transmitted via an electric communication line.

[0050] The information control unit 231 controls the input and output of information. For example, the information control unit 231 acquires teacher data from other devices (information processing devices and storage media) and records them in the teacher data storage unit 221. Also, for example, the information control unit 231 transmits the learned model stored in the learned model storage unit 223 to other devices (for example, the estimation device 30). The pre-processing control unit 232 converts the teacher data into a data format that can be input to the learned model. For example, when the 7-day strength of multiple specimens is included for the same type of blending case, the pre-processing control unit 232 may convert the 7-day strength of multiple specimens into an average of the 7-day strengths of the specimens. Similarly, when the 28-day strength of multiple specimens is included for the same type of blending case, the pre-processing control unit 232 may convert the 28-day strength of multiple specimens into an average of the 28-day strengths of the specimens. Also, the pre-processing control unit 232 may delete the 7-day strength of a specimen that is determined to show an abnormal value from the teacher data among the 7-day strengths of multiple specimens for the same type of blending case. Similarly, the pre-processing control unit 232 may delete from the training data the 28-day strength of a test specimen that is determined to show an abnormal value among the 28-day strengths of multiple test specimens for the same type of blending case.

[0051] The learning control unit 233 executes a learning process using the teacher data or preprocessed teacher data stored in the preprocessed teacher data storage unit 222. The learning control unit 233 generates a trained model for outputting an estimated value of the 28-day strength of each of the nine types of mix cases of the input unit cement amount and water content based on input information including the type of cement to be input, the unit cement amount and water content of each of the nine types of mix cases, and the 7-day strength of one or more mix cases, for example, by performing supervised learning. Hereinafter, among the nine types of mix cases, a mix case that is a target for inputting the 7-day strength to the estimation model is described as a strength input target mix case. The strength input target mix cases can be increased up to nine types, such as CASE2-2 for one type, CASE2-2 and 2-3 for two types, CASE2-1, 2-2 and 2-3 for three types, and CASE2-1, 2-2, 2-3 and 3-2 for four types. The learning control unit 233 records the generated trained model in the trained model storage unit 223. The trained model obtained by the learning control unit 233 may be transmitted to the estimation device 30 and recorded in the trained model storage unit 321 of the estimation device 30.

[0052] For example, the ExtraTreeRegressor of Python's scikitlearn can be used as the strength estimation model algorithm (see Reference 1, for example). ExtraTreeRegressor is characterized by being highly random and resistant to noise and outliers. In the strength estimation model, ExtraTreeRegressor performs regression by dividing the blending test data into multiple decision trees and averaging the values ​​predicted by each decision tree. Specifically, nine AI (artificial intelligence) models were created because the objective variable was the 28-day strength results of nine blending cases, and seven models were created for each model to perform cross-validation for this evaluation in seven divisions. Therefore, 9 x 7 = 63 AI models were created, and these 63 AI models were combined to form one strength estimation model. The 28-day strength results of the nine blending cases are the measurement results of the 28-day strength of each blending case in the blending test shown by the training data.

[0053] (References 1) Pierre Geurts et al., "Extremely randomized trees," Machine Learning 63(1), 3-42, 2006

[0054] 16 is a flowchart showing a specific example of processing of the learning device 20. First, the information control unit 231 acquires teacher data (step S101). The teacher data may be input by a user, acquired by communication from another information device, or acquired from a recording medium connected to the learning device 20. The preprocessing control unit 232 performs a predetermined preprocessing on the teacher data (step S102). The learning control unit 233 executes a learning process using the preprocessed teacher data, and records a learned model in the learned model storage unit 223 (step S103).

[0055] 17 is a schematic block diagram showing a specific example of a functional configuration of the estimation device 30. The estimation device 30 is configured using an information processing device such as a personal computer or a server device. The estimation device 30 includes a communication unit 31, a storage unit 32, and a control unit 33.

[0056] The communication unit 31 is a communication device. The communication unit 31 may be configured as, for example, a network interface. The communication unit 31 performs data communication with other devices via the network 70 according to the control of the control unit 33. The communication unit 31 may be a device that performs wireless communication or a device that performs wired communication.

[0057] The storage unit 32 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. The storage unit 32 stores data used by the control unit 33. The storage unit 32 may function as a trained model storage unit 321, for example.

[0058] The trained model storage unit 321 stores information about trained models that have been generated in advance by a learning process. Such a learning process may be executed by, for example, another device (for example, the learning device 20) or by the device itself (the estimation device 30).

[0059] The control unit 33 is configured using a processor such as a CPU and a memory. The control unit 33 functions as an information control unit 331, a preprocessing control unit 332, and an estimation unit 333 by the processor executing a program. All or part of the functions of the control unit 33 may be realized using hardware such as an ASIC, a PLD, or an FPGA. The above program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a semiconductor storage device (e.g., SSD), or a storage device such as a hard disk or a semiconductor storage device built into a computer system. The above program may be transmitted via an electric communication line.

[0060] The information control unit 331 acquires input information from other devices such as the terminal device 10. The input information includes information on the type of cement, the unit cement amount and water content of each of the nine types of mixing cases, and the 7-day strength of the strength input target mixing case and one or more specimens of the strength input target mixing case. The information on the strength input target mixing case is information indicating which of the nine types of mixing cases it is, that is, which of CASE1-1 to CASE3-3 it is. The information control unit 331 transmits information indicating the estimation result obtained by the estimation unit 333 to other devices such as the terminal device 10. Such exchange of information between the information control unit 331 and other devices may be performed, for example, by communication using the communication unit 31.

[0061] The preprocessing control unit 332 executes a predetermined preprocessing on the input information acquired by the information control unit 331. That is, the preprocessing control unit 332 converts the input information into a data format that can be input to the trained model by the same preprocessing as the preprocessing control unit 232 of the learning device 20. For example, when the 7-day strength of multiple specimens for the same type of blending case is included, the preprocessing control unit 332 may convert the 7-day strength into an average of the 7-day strengths of the specimens. In addition, the preprocessing control unit 332 may delete the 7-day strength of a specimen determined to show an abnormal value from the input information among the 7-day strengths of multiple specimens for the same type of blending case.

[0062] The estimation unit 333 performs estimation processing using the trained model stored in the trained model storage unit 321 and the input information whose data format has been converted by the preprocessing control unit 332. The estimation unit 333 may perform estimation processing using the trained model stored in the trained model storage unit 321 and the input information acquired by the information control unit 331. The estimation processing outputs estimated values ​​of 28-day strength for each of the nine types of blending cases included in the input information.

[0063] Next, the estimation of 28-day strength using the estimation device 30 will be described. First, the tester considers the type of cement and nine types of mix cases as shown in FIG. 1 for the sample (soil or sand) to be subjected to mix testing. Depending on the number of strength input mix cases, the tester decides which mix cases from CASE 1-1 to 3-3 to at least perform mix testing for. For the mix cases 1 to 9 that have been decided to be subjected to mix testing, the tester performs steps S1 to S8 of the mix testing shown in FIG. 2 to obtain the 7-day strength.

[0064] FIG. 18 is a flowchart showing a specific example of the processing of the estimation device 30. The information control unit 331 acquires input information from the terminal device 10 (step S201). The input information includes the type of cement used in the blending test, the unit cement amount and water content of the nine types of blending cases examined, the strength input blending case, and the 7-day strength of each test specimen of the strength input blending case obtained by the blending test. The pre-processing control unit 332 performs pre-processing on the input information to convert it into a data format that can be input to the trained model (step S202). The estimation unit 333 performs estimation processing using the input information whose data format has been converted (step S203). The estimation unit 333 transmits information on the estimated value of the 28-day strength of each of the nine types of blending cases obtained as the estimation result to the terminal device 10 (step S204).

[0065] In addition, when the tester refers to the estimated result output in step S204 and determines that it is necessary to obtain the 28-day strength of a different mix case, he / she uses the terminal device 10 to generate input information in which the unit cement amount and water content of the mix case are changed. For example, the strength input mix case is one type, CASE2-2. In this case, input information can be generated in which the unit cement amount of CASE1 and CASE3 and the water content values ​​of CASE1-1, 1-2, 1-3, 2-1, 2-3, 3-1, 3-2, and 3-3 are changed. In addition, the strength input mix case is four types, CASE2-1, 2-2, 2-3, and 3-2. In this case, input information can be generated in which the unit cement amount of CASE1 and the water content values ​​of CASE1-1, 1-2, 1-3, 3-1, and 3-3 are changed. In addition, if the strength input mix case is one type, CASE2-2, and the 7-day strength of multiple mix cases is measured in the mix test, it is also possible to generate input information using each of those mix cases as the strength input mix case of CASE2-2. The estimation device 30 can obtain the 28-day strength of erosion control soil cement with unit cement amount and water content that was not subjected to mix tests without performing further mix tests, by acquiring the generated input information from the terminal device 10 and performing the processing of Fig. 18.

[0066] FIG. 19 is a diagram showing an example of an input screen displayed on the terminal device 10. In the input screen, the type of cement (E1), the unit cement amount (E2) and water content (E3) of each mix case, and the 7-day strength (E4) of each specimen of the strength input mix case are input. The strength input mix case among each mix case is identified by the mix case number at the beginning of the row in which the 7-day strength (E4) is input. FIG. 19 shows an example in which there are three types of strength input mix cases, CASE2-1, 2-2, and 2-3. For example, when there is one type of strength input mix case, the 7-day strength of CASE2-2 is input, and the 7-day strengths of CASE2-1 and 2-3 are not input. The information control unit 331 of the estimation device 30 acquires the input information input on the input screen (step S201 in FIG. 18). When the 7-day strengths of multiple specimens are input for one strength input blending case, the preprocessing control unit 332 performs preprocessing to obtain the average (step S202 in FIG. 18). The input screen shown in FIG. 19 displays the average of the 7-day strengths after preprocessing (E5). The terminal device 10 may set the average of the 7-day strengths as the input information. The estimation unit 333 transmits the 28-day strengths of each blending case obtained by inputting the preprocessed input information into the trained model to the terminal device 10 (step S203 in FIG. 18). The terminal device 10 displays the received 28-day strengths (E6) of each blending case.

[0067] In addition, in step S201 of Figure 18, you may input the average of the 7-day strength (E5) instead of inputting the 7-day strength (E4). In this case, you do not need to perform pre-processing to obtain the average for each strength input blend case in step S202.

[0068] In the input screen shown in Figure 19, by changing the cement amount of the cement case to which the strength input mix case does not belong and the water content of the mix case other than the strength input mix case, it is possible to obtain an estimate of the 28-day strength of other unit cement amounts and water content. Also, when the 7-day strength of multiple mix cases is measured in a mix test, the unit cement and water content of each mix case may be set and the input information changed so that the 7-day strength of any of those mix cases becomes the strength input mix case of CASE2-2.

[0069] In the estimation system 100 configured as described above, it is possible to estimate the 28-day strength of multiple types of blending cases by using information on the 7-day strength of at least one specimen of at least one type of blending case. Specifically, it is as follows. The estimation system 100 executes a learning process by machine learning using blending test data of the 7-day strength and 28-day strength of multiple blending cases obtained as a result of the blending test, and obtains a trained model. Then, the estimation system 100 acquires new input information on the 7-day strength of the blending test performed on at least some of the multiple blending cases to be estimated for the 28-day strength, and acquires estimated information on the 28-day strength of each of the multiple blending cases to be estimated based on the acquired input information.

[0070] Although the above describes a case where there are nine types of combination cases, the types of combination cases may be other than nine. In addition, training data generated based on the input information input to the estimation device 30 in the process of FIG. 18 and the estimation result obtained using the input information may be used in the learning process of FIG. 16.

[0071] FIG. 20 is a diagram showing an outline of a hardware configuration example of an information processing device 90 applied to the present embodiment. The information processing device 90 includes a processor 91, a main storage device 92, a communication interface 93, an auxiliary storage device 94, an input / output interface 95, and an internal bus 96. The processor 91, the main storage device 92, the communication interface 93, the auxiliary storage device 94, and the input / output interface 95 are communicably connected to each other via the internal bus 96. The information processing device 90 may be applied to, for example, the learning device 20 and the estimation device 30. In this case, for example, the communication unit 21 and the communication unit 31 may be configured using the communication interface 93. For example, the memory unit 22 and the memory unit 32 may be configured using the auxiliary storage device 94. In addition, the control unit 23 and the control unit 33 may be configured using the processor 91 and the main storage device 92.

[0072] (Modification) In this embodiment, the terminal device 10 and the estimation device 30 are configured as different devices, but may be configured as an integrated device. FIG. 21 is a diagram showing a modified example of the estimation device 30 configured in this way. The estimation device 30 shown in FIG. 21 includes an input unit 34 and an output unit 35. The input unit 34 and the output unit 35 of the estimation device 30 shown in FIG. 21 function similarly to the input unit 12 and the output unit 13 of the terminal device 10, respectively. The control unit 33 operates in response to an operation on the input unit 34, performs an estimation process using the input information that has been input, and outputs information using the output unit 35.

[0073] In this embodiment, the learning device 20 and the estimation device 30 are configured as different devices, but may be configured as an integrated device. FIG. 22 is a diagram showing a modified example of the estimation device 30 configured in this manner. The storage unit 32 of the estimation device 30 shown in FIG. 22 also functions as a teacher data storage unit 322 and a preprocessed teacher data storage unit 323. The control unit 33 of the estimation device 30 shown in FIG. 22 also functions as a learning control unit 334. The teacher data storage unit 322 and the preprocessed teacher data storage unit 323 function in the same manner as the teacher data storage unit 221 and the preprocessed teacher data storage unit 222 of the learning device 20, respectively. The preprocessing control unit 332 executes not only the preprocessing of the estimation device 30 (preprocessing of input information) but also the preprocessing of the learning device 20 (preprocessing of teacher data). The learning control unit 334 functions in the same manner as the learning control unit 233 of the learning device 20.

[0074] Furthermore, even when the learning device 20 and the estimation device 30 are configured as different devices, the estimation device 30 may use the learned model storage unit 223 of the learning device 20 as the learned model storage unit 321.

[0075] The learning device 20 may be implemented using a plurality of information processing devices. For example, the learning device 20 may be implemented using a device such as a cloud. For example, in the learning device 20, the storage unit 22 and the control unit 23 may be implemented in different information processing devices. For example, the storage unit 22 of the learning device 20 may be distributed and implemented in a plurality of information processing devices. The estimation device 30 may be implemented using a plurality of information processing devices. For example, the estimation device 30 may be implemented using a device such as a cloud. For example, in the estimation device 30, the storage unit 32 and the control unit 33 may be implemented in different information processing devices. For example, the storage unit 32 of the estimation device 30 may be distributed and implemented in a plurality of information processing devices.

[0076] Furthermore, the learning control unit 233 of the learning device 20 and the learning control unit 334 of the estimation device 30 may create a strength estimation model for each type of cement. In this case, the estimation unit 333 of the estimation device 30 selects a trained model based on information on the type of cement included in the input information, inputs the input information excluding the information on the type of cement to the selected trained model, and obtains an estimated value of the 28-day strength for each of the multiple blend cases.

[0077] An evaluation experiment of the accuracy of prediction using the trained model of this embodiment will be described. The trained model used was an ExtraTreeRegressor. Two trained models were trained by the learning device 20. One used 3034 pieces of training data, and the other used 2178 pieces of training data. Based on the data from the evaluation test performed according to the procedure shown in Figure 2, input information was generated in which the number of blend cases for strength input was increased from one to nine. The estimation device 30 inputted the input information to each trained model to obtain an estimation result of the 28-day strength for each of the nine blend cases. The coefficient of determination R 2 For each trained model, the coefficient of determination R 2 was 0.8 or higher, and the accuracy increased as the number of mix cases for which strength was entered increased.

[0078] FIG. 23 is a diagram showing the 28-day strength of a conventional mix test that was actually performed and the estimated results of the 28-day strength using this embodiment. The prediction by this embodiment was performed for two patterns: one type of strength input mix case (hereinafter referred to as "one-case prediction result") and three types (hereinafter referred to as "three-case prediction result"). Symbols F1, F2, and F3 are the measured 28-day strength obtained by the conventional mix test, symbols G1, G2, and G3 are the one-case prediction result (AI_1), and symbols H1, H2, and H3 are the three-case prediction result (AI_3). Symbols F1, G1, and H1 are the cement amount of 50 (C-50), symbols F2, G2, and H2 are the cement amount of 100 (C-100), and symbols F3, G3, and H3 are the cement amount of 150 (C-150).

[0079] Coefficient of determination for single case prediction results (R 2 The value was 0.83, and the coefficient of determination (R 2 The value was 0.90. Although the prediction results for the three cases were more accurate, taking into account the variation in natural soil, the deviation in the prediction results for one case was very minor. As shown above, the 28-day strength estimates for nine types of mix cases were obtained with high estimation accuracy.

[0080] According to the embodiment described above, the minimum number of specimens to be prepared for physical mix tests can be reduced from the conventional 54 specimens (3 specimens for each of 9 cases) to 3 specimens for 1 case or 1 specimen for 1 case. In addition, the amount of soil and sand to be collected for mix tests is significantly reduced. In addition, the time required for the results of the mix tests to be known is shortened from the conventional 28 days to 7 days, so that the construction period is reduced. In addition, even if the required quality is not satisfied with the initial mix case, information useful for determining the recommended mix can be obtained without conducting additional mix tests by changing the mix case settings in the input information. Therefore, the risk of additional sample collection, increased construction period, and associated increased costs is reduced. In addition, if the mix test facility is far from the sample collection site or if it is difficult to transport the samples, the JIS standard test can be performed near the collection site, 3 specimens or 1 specimen for 1 mix case can be prepared on-site, and information on the 7-day strength can be obtained and used as input data, making it possible to consider the recommended mix. For example, a mix test kit containing one mix case's worth of specimens is easy to transport, so even if the site is overseas, it is possible to obtain the data necessary for the recommended mix in a short period of time by collecting about 15 kg of soil and sand at the site and measuring its strength after 7 days. The above explanation uses erosion control soil cement as an example, but it can also be applied to soil cement, which is a mixture of soil, solidifying material, and water.

[0081] According to the embodiment described above, the estimation system has a learning control unit and an estimation unit. The learning device may have the learning control unit, and the estimation device may have the estimation unit. The learning control unit uses teacher data indicating the solidification material amount and water content of each of the multiple types of mixing cases of soil cement, the first period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a first period, and the second period strength, which is the strength after a second period longer than the first period, to learn a strength estimation model by machine learning, which inputs the solidification material amount and water content of each of the multiple types of mixing cases and the first period strength of the soil cement of one or more mixing cases, and outputs the second period strength of the soil cement of each of the multiple types of mixing cases having the input solidification material amount and water content. The estimation unit inputs the solidification material amount and water content of each of the multiple types of mixing cases to be the second period strength estimation target, and the first period strength of the soil cement of one or more mixing cases among the multiple types of mixing cases to be the second period strength estimation target, to the learned strength estimation model, thereby obtaining an output of the second period strength of the soil cement of each of the multiple types of mixing cases having the input solidification material amount and water content. For example, the first period is 7 days in the above embodiment, and the second period is 28 days in the above embodiment. The first period is a period during which the hydration reaction of the cement has not progressed, and the second period is a period during which it is possible to confirm whether the hydration reaction of the cement has progressed to a strength that ensures that the material has hardened.

[0082] The training data may further include information on the type of solidification material. The strength estimation model further inputs the type of solidification material. The estimation unit inputs the type of solidification material, the amount of solidification material and water content of each of the multiple types of mix cases, and the first period strength of the soil cement of one or more mix cases to the strength estimation model, thereby obtaining an output of the second period strength of the soil cement of each of the multiple types of mix cases using the input type of solidification material.

[0083] In addition, in the above embodiment, an example was described in which the second period strength is higher than the first period strength, but it is also applicable to the case of soil cement in which the second period strength is lower than the first period strength.

[0084] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included. [Explanation of symbols]

[0085] 100... estimation system, 10... terminal device, 11... communication unit, 12... input unit, 13... output unit, 14... memory unit, 15... control unit, 20... learning device, 21... communication unit, 22... memory unit, 221... teacher data memory unit, 222... preprocessed teacher data memory unit, 223... trained model memory unit, 23... control unit, 231... information control unit, 232... preprocessing control unit, 233... learning control unit, 30... estimation device, 31... communication unit, 32... memory unit, 321... trained model memory unit, 322... teacher data memory unit, 323... preprocessed teacher data memory unit, 33... control unit, 331... information control unit, 332... preprocessing control unit, 333... estimation unit, 334... learning control unit, 34... input unit, 35... output unit

Claims

1. a learning control unit that uses teacher data indicating the amount of solidifying agent and the water content of each of a plurality of types of mixing cases of soil cement, and a first period strength which is the strength of the soil cement of each of the plurality of types of mixing cases after a first period has elapsed and a second period strength which is the strength of the soil cement of each of the plurality of types of mixing cases after a second period longer than the first period has elapsed, to learn a strength estimation model by machine learning, which receives as input the amount of solidifying agent and the water content of each of a plurality of types of mixing cases and the first period strength of the soil cement of one or more of the mixing cases, and outputs the second period strength of the soil cement of each of the plurality of types of mixing cases having the input amount of solidifying agent and the water content; A learning device comprising:

2. The teaching data further includes information on the type of solidification material, The learning control unit inputs the type of solidification material, the amount of solidification material and the water content of each of the multiple types of mixing cases, and the first period strength of the soil cement of one or more of the mixing cases, and learns a strength estimation model by machine learning to output a second period strength of the soil cement of each of the multiple types of mixing cases having the input amount of solidification material and the water content. The learning device according to claim 1 .

3. an estimation unit which obtains an output of a second period strength of the soil cement for each of the plurality of mix cases having the inputted solidification material amount and water content, by inputting the solidification material amount and water content of each of the plurality of mix cases for which a second period strength is to be estimated and the first period strength of the soil cement for one or more mix cases among the plurality of mix cases for which a second period strength is to be estimated, into a strength estimation model which has been trained using teacher data indicating the solidification material amount and water content of each of the plurality of mix cases of soil cement, and a first period strength which is the strength of the soil cement for each of the plurality of mix cases after a first period has elapsed and a second period strength which is the strength after a second period longer than the first period has elapsed; An estimation device comprising:

4. The teaching data further includes information on the type of solidification material, The estimation unit inputs, into the strength estimation model, the type of solidification material, the amount of solidification material and the water content of each of the multiple types of mix cases for which second period strength is estimated, and the first period strength of the soil cement of one or more of the multiple types of mix cases for which second period strength is estimated, thereby obtaining an output of the second period strength of the soil cement of each of the multiple types of mix cases having the input amount of solidification material and the water content. The estimation device according to claim 3 .

5. a learning control unit that uses teacher data indicating the amount of solidifying agent and water content of each of multiple types of mixing cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a first period has elapsed, and a second period strength, which is the strength of the soil cement of each of the multiple types of mixing cases after a second period longer than the first period has elapsed, to learn a strength estimation model by machine learning, which inputs the amount of solidifying agent and water content of each of the multiple types of mixing cases and the first period strength of the soil cement of one or more of the mixing cases, and outputs the second period strength of the soil cement of each of the multiple types of mixing cases having the input amount of solidifying agent and water content; an estimation unit that inputs, into the strength estimation model, the amount of solidifying agent and the water content of each of the multiple types of mix cases that are the subject of second-period strength estimation, and the first-period strength of the soil cement of one or more of the multiple types of mix cases that are the subject of second-period strength estimation, and obtains an output of the second-period strength of the soil cement of each of the multiple types of mix cases that have the input amount of solidifying agent and the water content; An estimation system comprising:

6. a learning step of learning, by machine learning, a strength estimation model that uses teacher data indicating the amount of solidifying agent and water content of each of multiple types of mix cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mix cases after a first period has elapsed, and a second period strength, which is the strength of the soil cement of each of the multiple types of mix cases after a second period longer than the first period has elapsed, by inputting the amount of solidifying agent and water content of each of the multiple types of mix cases and the first period strength of the soil cement of one or more of the mix cases, and outputting the second period strength of the soil cement of each of the multiple types of mix cases having the input amount of solidifying agent and water content; A learning method that has

7. an estimation step of inputting the amount of solidifying agent and the water content of each of the multiple types of mix cases that are the subject of second period strength estimation and the first period strength of the soil cement of one or more mix cases among the multiple types of mix cases that are the subject of second period strength estimation into a strength estimation model trained using teacher data indicating the amount of solidifying agent and the water content of each of the multiple types of mix cases of soil cement, and a first period strength, which is the strength of the soil cement of each of the multiple types of mix cases after a first period has elapsed, and a second period strength, which is the strength after a second period that is longer than the first period has elapsed, thereby obtaining an output of the second period strength of the soil cement of each of the multiple types of mix cases that have the input amount of solidifying agent and the water content; The estimation method has the following structure:

8. Computer, A program for causing the learning device according to claim 1 or 2 to function.

9. Computer, A program for causing the estimation device according to claim 3 or 4 to function.

Citation Information

Patent Citations

  • Early determination method for strength of erosion control soil cement and manufacturing method for erosion control soil cement

    JP2021009062A

Cited By

  • Soil cement mix design method and soil cement construction method

    JP2026165047A

  • Soil cement mix design method and soil cement construction method

    JP7732116B1