JP6857167B2 - Stratigraphy determination device and program - Google Patents

Description

The present invention relates to a stratigraphy determination device and a program.

As a technique for investigating the ground, there is a Swedish sounding test (see Non-Patent Document 1). After the ground survey by such a technique, the stratigraphy of the stratum may be determined based on the result of the ground survey. The stratigraphy of the stratum is determined based on the results of the ground survey by the ground engineer based on his own knowledge and experience. However, since the stratigraphy is judged by the ground engineer based on his / her own knowledge and experience, the judgment result may vary from one ground engineer to another. In addition, since the ground engineer determines the stratigraphy based on his / her own knowledge and experience, the result of the stratigraphy determination may be incorrect depending on the technique of the ground engineer.
Due to such factors, the accuracy of the determination result of the stratigraphy of the stratum based on the result of the ground survey has been low in the past.

"Method for Swedish weight sounding test" JIS A 1221: 2013

In view of the above circumstances, it is an object of the present invention to provide a technique for improving the accuracy of the determination result of stratigraphy.

In one aspect of the present invention, information indicating the position dependence of the physical quantity of the soil quality in the depth direction is used as the soil quality information, and a plurality of predetermined sets that classify the soil quality with the soil quality information. as related information information indicating a relationship between the soil aggregate is, on the basis of said soil information entered with said relationship information, it determines stratigraphy in the depth direction of the land showing the soil information entered The second determination unit includes a first determination unit and a second determination unit that determines the stratum in the depth direction of the land based on the input soil information and the determination result of the first determination unit. In the determination unit, the input soil information and the information indicating the soil set to which the soil at each position in the depth direction indicated by the input soil information belongs are determined to have a thickness in the depth direction of the land. If there is a thin layer that is less than or equal to the thickness of, and it indicates that the thin layer is not located at either the shallowest position or the deepest position in the depth direction, the thin layer is used for filling. It is determined whether or not the thin layer is sandwiched between the filling soils, and if the thin layer is sandwiched between the filling soils, it is determined that the soil aggregate to which the thin layer belongs is the filling soil, and the thin layer is not sandwiched between the filling soils. In this case, it is a stratum determination device that determines that the soil set to which the thin layer belongs is the same as the soil set to which the stratum located above the thin layer belongs.

One aspect of the present invention is the stratification determination device, wherein the first determination unit is based on a learning model learned by machine learning using the relational information as teacher data and the input soil information. , Determine the stratification in the depth direction of the land indicated by the input soil information.

One aspect of the present invention is the above-mentioned stratigraphy determination device, and the first determination unit further determines the stratigraphy based on the information of GIS (Geographic Information System).

One aspect of the present invention is the above-mentioned stratigraphy determination device, in which the physical quantities relating to the soil quality of the land are the load, the half rotation speed, and the penetration depth in the Swedish sounding test.

One aspect of the present invention is a program for operating a computer as the above-mentioned stratigraphy determination device.

INDUSTRIAL APPLICABILITY According to the present invention, it is possible to provide a technique for improving the accuracy of the determination result of stratigraphy.

It is a figure which shows an example of the system structure of the stratigraphy determination system 100 of an embodiment. It is a figure which shows an example of the input soil information in an embodiment. It is a flowchart which shows an example of the process flow according to the stratigraphic algorithm executed by the 2nd determination unit 232 in embodiment. It is a figure which shows an example of the flow of the process executed by the stratigraphy determination apparatus 2 of an embodiment. It is a figure which shows the determination result of the 1st determination unit 231 output by the output unit 24 in embodiment. It is a figure which shows the determination result of the 2nd determination unit 232 output by the output unit 24 in embodiment.

(Embodiment)
FIG. 1 is a diagram showing an example of a system configuration of the stratigraphy determination system 100 of the embodiment. Hereinafter, information indicating the positional dependence of the physical quantity on the soil quality of the land in the depth direction is referred to as soil quality information. Hereinafter, the physical quantity related to soil quality is referred to as soil physical quantity. The soil physical quantity is, for example, a physical quantity obtained by a geological survey. The physical quantities obtained by the geological survey are, for example, the physical quantities obtained by the Swedish sounding test, such as load, half rotation speed, penetration depth, penetration amount, half rotation speed per meter, converted N value, and allowable support. It is a force qa or the like.

The stratigraphy determination system 100 determines the stratigraphy in the depth direction of the land indicated by the input soil information based on the input soil information (hereinafter referred to as "input soil information"). Specifically, in the stratigraphy determination system 100, the soil quality at each position belongs to a plurality of predetermined sets for each position in the depth direction indicated by the input soil quality information, among a plurality of sets for classifying the soil quality. Determine the set. That is, in the stratigraphy determination system 100, the soil quality at each position to be determined is a predetermined set of a plurality of sets for classifying the soil quality for each position in the depth direction indicated by the input soil quality information. It is determined which set it belongs to.
Hereinafter, a plurality of predetermined sets that classify soil quality are referred to as soil quality sets. The soil set may be, for example, a set of soils defined by the stratum name such as embankment, sandy soil, volcanic ash soil, and cohesive soil.

The stratigraphy determination system 100 includes a machine learning unit 1 and a stratigraphy determination device 2.
The machine learning unit 1 uses machine learning as teacher data, which is information indicating the relationship between the soil information and the soil set, to obtain the soil information and the soil set to which the soil at each position in the depth direction indicated by the soil information belongs. A learning model (hereinafter referred to as "classification model") that expresses the relationship between

The machine learning unit 1 is configured by using an information processing device such as a personal computer or a server. The machine learning unit 1 includes a CPU (Central Processing Unit), a memory, an auxiliary storage device, and the like connected by a bus, and executes a program. The machine learning unit 1 functions as a device for learning a classification model by machine learning by executing a program.
The machine learning unit 1 outputs a classification model of the learning result to the stratigraphy determination device 2.

The stratigraphy determination device 2 acquires the input soil quality information, and determines the soil quality set to which the soil quality at each position in the depth direction indicated by the input soil quality information belongs, based on the classification model which is the learning result of the machine learning unit 1.
The stratigraphy determination device 2 is configured by using an information processing device such as a personal computer or a server. The stratigraphy determination device 2 includes a CPU (Central Processing Unit), a memory, an auxiliary storage device 21, and the like connected by a bus, and executes a program. The stratigraphy determination device 2 functions as a device including an input unit 22, a control unit 23, and an output unit 24 by executing a program. The CPU and the memory function as the control unit 23 by executing the program stored in the auxiliary storage device 21.

The auxiliary storage device 21 is configured by using a storage device such as a magnetic hard disk device or a semiconductor storage device. The auxiliary storage device 21 stores a classification model that is a learning result learned by the machine learning unit 1.

The input unit 22 includes an input device such as a mouse, a keyboard, and a touch panel. The input unit 22 may be configured as an interface for connecting these input devices to its own device. The input unit 22 receives the input of the classification model for the own device. The classification model input to the input unit 22 is recorded in the auxiliary storage device 21 by the recording unit (not shown).
The input unit 22 accepts the input of the input soil information. The input unit 22 outputs the input input soil information to the control unit 23.

The control unit 23 includes a first determination unit 231 and a second determination unit 232.
The first determination unit 231 determines the soil set to which the soil at each position in the depth direction indicated by the input soil information belongs, based on the input soil information and the classification model stored in the auxiliary storage device 21. The determination result of the first determination unit 231 is output to the second determination unit 232 and the output unit 24.

The second determination unit 232 determines the stratigraphy by the stratigraphy algorithm, which is an algorithm related to stratigraphy, based on the input soil information and the determination result of the first determination unit 231. The stratigraphic algorithm may be any algorithm as long as it is an algorithm related to stratigraphy. The stratigraphic algorithm may be, for example, an algorithm based on the law of formation identification. The details of the stratigraphic algorithm will be described later. The determination result of the second determination unit 232 is output to the output unit 24.

The output unit 24 outputs the determination result of the first determination unit 231 and the determination result of the second determination unit 232. The output unit 24 may output the determination result of the first determination unit 231 and the determination result of the second determination unit 232. The output unit 24 may include, for example, a display device such as a CRT (Cathode Ray Tube) display, a liquid crystal display, or an organic EL (Electro-Luminescence) display. The output unit 24 may be configured as, for example, an interface for connecting these display devices to its own device.

FIG. 2 is a diagram showing an example of input soil information in the embodiment.
The input soil information shown in FIG. 2 includes the load (km), half rotation speed (Na), penetration amount (cm), half rotation speed per meter (Nsw), converted N value, and allowable value obtained by the Swedish sounding test. It shows the penetration depth (m) dependence of the bearing capacity. The penetration depth (m) indicates the depth.

FIG. 3 is a flowchart showing an example of a processing flow according to a stratigraphic algorithm executed by the second determination unit 232 in the embodiment.

The second determination unit 232 acquires the input soil information and the determination result of the first determination unit 231 (step S101). The second determination unit 232 determines whether or not there is a stratum having a thickness equal to or less than a predetermined thickness (hereinafter referred to as "thin layer") based on the input soil information and the determination result of the first determination unit 231. (Step S102). The predetermined thickness is, for example, 25 cm.
When there is a thin layer (step S102: YES), the second determination unit 232 determines whether or not the thin layer is located at either the shallowest position in the depth direction or the deepest position in the depth direction. Is determined (step S103).
When the thin layer is not located at any position (step S103: NO), the second determination unit 232 determines whether or not the thin layer is sandwiched between the embankments (step S104).
When the thin layer is sandwiched between the embankments (step S104: YES), the second determination unit 232 determines that the soil aggregate to which the thin layer belongs is the embankment (step S105).
After step S105, the second determination unit 232 ends the process.

On the other hand, when the thin layer is not sandwiched between the embankments (step S104: NO), the second determination unit 232 sets the soil set to which the thin layer belongs as the soil set of the soil located above the thin layer in the depth direction. It is determined that they are the same (step S106).
After step S106, the second determination unit 232 ends the process.

On the other hand, in step S103, when the thin layer is located at any position (step S103: YES), the second determination unit 232 ends the process.
On the other hand, in step S102, when there is no thin layer (step S102: NO), the second determination unit 232 ends the process.

In this way, when the determination result of the first determination unit 231 is a result contrary to the theory of stratigraphy, the second determination unit 232 corrects the result so as not to violate the theory of stratigraphy.

FIG. 4 is a diagram showing an example of a flow of processing executed by the stratigraphy determination device 2 of the embodiment.
The input unit 22 acquires the input soil information (step S201).
The first determination unit 231 determines the soil set to which the soil at each position in the depth direction indicated by the input soil information belongs based on the input soil information and the classification model stored in the auxiliary storage device 21 (step S202). ..
The second determination unit 232 determines the stratigraphy by the stratigraphy algorithm based on the input soil information and the determination result of the first determination unit 231 (step S203).
The output unit 24 outputs the determination result of step S202 or step S203 (step S204).
After step S204, the stratigraphy determination device 2 ends the process.

Here, FIGS. 5 and 6 show the determination result of the first determination unit 231 and the determination result of the second determination unit 232, and the determination result of the second determination unit 232 is the determination result of the first determination unit 231. Show different.

FIG. 5 is a diagram showing a determination result of the first determination unit 231 output by the output unit 24 in the embodiment.
In the graph of FIG. 5, the vertical axis represents the depth. The vertical axis indicates that the value is large and the position is deep. In the graph of FIG. 5, the horizontal axis represents the magnitude of a predetermined soil physical quantity. In FIG. 5, the judgment results of the first judgment unit 231 are, in order from the shallowest depth, "filling / buried soil (sandy soil)", "volcanic ash soil", "cohesive soil (sand layer mixture)", and " Indicates that it is "volcanic ash soil", "sandy soil", "cohesive soil", "sandy soil", "cohesive soil (mixed sand layer)", "sandy soil", "cohesive soil (mixed sand layer)". ..

FIG. 6 is a diagram showing a determination result of the second determination unit 232 output by the output unit 24 in the embodiment.
In the graph of FIG. 6, the vertical axis represents the depth. The vertical axis indicates that the value is large and the position is deep. In the graph of FIG. 6, the horizontal axis represents the magnitude of a predetermined soil physical quantity. In FIG. 6, the judgment results of the second judgment unit 232 are "filling / buried soil (sandy soil)", "volcanic ash soil", "sandy soil", and "cohesive soil" in order from the shallowest depth. Sand layer mixture) ”.

FIG. 5 shows that the determination result of the first determination unit 231 is that there is cohesive soil (sand layer mixture) between the volcanic ash soil and the volcanic ash soil. However, according to the theory of stratigraphy, such a stratigraphy is not possible. On the other hand, FIG. 6 shows that there is no cohesive soil (sand layer mixture) between the volcanic ash soil and the volcanic ash soil in the determination result of the second determination unit 232. In this way, the second determination unit 232 corrects the determination result of the first determination unit 231 that violates the theory of stratigraphy, and acquires the determination result of the stratigraphy that does not violate the theory of stratigraphy.

Since the stratigraphy determination system 100 configured in this way includes the first determination unit 231 and the second determination unit 232, it outputs a stratigraphy determination result that does not violate the stratigraphy theory based on the input soil information. be able to. Therefore, it is possible to suppress an increase in the labor of engineers involved in determining the type of stratum.

(Modification example)
The classification model may be a learning result based on the physical quantity acquired by the geological survey and the information of GIS (Geographic Information System).

The first determination unit 231 is not necessarily based on the classification model and the input soil information, based on the relationship between the soil information and the soil set to which the soil at each position in the depth direction indicated by the soil information belongs and the input soil information. It is not necessary to determine the soil set to which the soil at each position in the depth direction indicated by the input soil information belongs.

In the stratification determination system 100, each of the depth directions indicated by the input soil information is based on a learning model representing the relationship between the input soil information and the soil set to which the soil at each position in the depth direction indicated by the input soil information belongs. The machine learning unit 1 does not necessarily have to be provided as long as it is possible to determine the soil set to which the soil at the position belongs.

The stratigraphy determination device 2 does not necessarily have to include the first determination unit 231. In the stratigraphy determination device 2 not provided with the first determination unit 231, the second determination unit 232 stratigraphy based on the input soil information and the information indicating the soil set to which the soil at each position in the depth direction indicated by the input soil information belongs. To judge.

The soil information may include topographical information. Topographical information is information about topography. The terrain information may be, for example, longitude or latitude. The terrain information may be altitude. The terrain information may be, for example, a terrain classification name. The topographical information may be, for example, information on surrounding active faults.
The soil information may include soil information. Soil information is information about soil. The soil information may be, for example, a soil name.

In addition to the soil quality information, map information may be further input to the stratigraphy determination system 100. The map information is information including topographical information and soil information. Hereinafter, information including soil information and map information is referred to as ground information. The stratigraphy determination system 100 of such a modified example determines the stratigraphy in the depth direction of the land indicated by the input soil information based on the ground information. Specifically, the stratification determination system 100 of such a modified example learns a ground learning model in the machine learning unit 1 by machine learning using information indicating the relationship between the ground information and the soil set as teacher data. The ground learning model is a learning model that represents the relationship between the ground information and the soil set to which the soil at each position in the depth direction indicated by the soil information belongs. Based on the ground learning model, the stratigraphy determination system 100 of such a modified example determines the soil set to which the soil quality at each position in the depth direction indicated by the input ground information belongs by the stratigraphy determination device 2.

All or part of each function of the machine learning unit 1 and the stratification determination device 2 uses hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), and an FPGA (Field Programmable Gate Array). May be realized. The program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a flexible disk, a magneto-optical disk, a portable medium such as a ROM or a CD-ROM, or a storage device such as a hard disk built in a computer system. The program may be transmitted over a telecommunication line.

The stratigraphy determination device 2 may be implemented by using a plurality of information processing devices connected so as to be able to communicate via a network. In this case, each functional unit included in the stratigraphy determination device 2 may be distributed and mounted in a plurality of information processing devices. For example, the first determination unit 231 and the second determination unit 232 may be mounted on different information processing devices.
Further, the machine learning unit 1 and the stratigraphy determination device 2 do not necessarily have to be mounted in different housings, and may be mounted in one housing.

Although the embodiments of the present invention have been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs and the like within a range that does not deviate from the gist of the present invention.

100 ... stratigraphy determination system, 1 ... machine learning unit, 2 ... stratigraphy determination device, 21 ... auxiliary storage device, 22 ... input unit, 23 ... control unit, 231 ... first determination unit, 232 ... second determination unit, 24 ... Output unit

Claims (5)

The information indicating the position dependence of the physical quantity of the soil quality in the depth direction is used as the soil quality information, and the relationship between the soil quality information and the soil quality set, which is a plurality of predetermined sets for classifying the soil quality. as related information information indicating, said based on the relationship information and the soil information input, a first determination unit that determines stratigraphy in the depth direction of the land showing the soil information input,
A second determination unit that determines the stratigraphy in the depth direction of the land based on the input soil information and the determination result of the first determination unit.
With
In the second determination unit, the input soil information and the information indicating the soil set to which the soil at each position in the depth direction indicated by the input soil information belongs are thick in the depth direction of the land. When there is a thin layer whose thickness is less than or equal to a predetermined thickness, and it indicates that the thin layer is not located at either the shallowest position or the deepest position in the depth direction, the thin layer Is sandwiched between the filling soils, and if the thin layer is sandwiched between the filling soils, it is determined that the soil aggregate to which the thin layers belong is the filling soil, and the thin layers are sandwiched between the filling soils. If not, it is determined that the soil set to which the thin layer belongs is the same as the soil set to which the layer located above the thin layer belongs.
Stratigraphy determination device. The first determination unit is a layer in the depth direction of the land indicated by the input soil information based on the learning model learned by machine learning using the relational information as teacher data and the input soil information. Judging the order,
The stratigraphy determination device according to claim 1. The first determination unit further determines the stratigraphy based on the information of GIS (Geographic Information System).
The stratigraphy determination device according to claim 1 or 2. The physical quantities relating to the soil quality of the land are the load, half rotation speed, and penetration depth in the Swedish sounding test.
The stratigraphy determination device according to any one of claims 1 to 3. A program for operating a computer as the stratigraphy determination device according to any one of claims 1 to 4.

Images