Evaluation system, learning model creation device, evaluation method, learning model creation method, and program

Abstract

To provide an evaluation system, a learning model creation device, an evaluation method, a learning model creation method, and a program that can estimate feed conditions of concrete.SOLUTION: An evaluation system comprises: a reception part which receives pressure data detected by a pressure sensor arranged in a feed path of concrete in a transport pipe and detection time information on the time when the pressure sensor detects the pressure data; and an estimation part which generates time-series data of the pressure data based upon the received pressure data and detection time information, and estimates the feed state of the concrete being fed in the transport pipe based upon the generated time-series data of the pressure data, and the relation between the time-series data of the pressure data in the feed path of the concrete and the feed state of the concrete.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] Concrete with a small slump or highly viscous, high-flow, high-strength concrete generally experiences large pressure loss in the pipes. Therefore, if the pumping distance is long or the pumping height is high, the pumping load on the pump increases and pumping tends to become more difficult. Even if the mixture is carefully selected, variations in quality during actual production can lead to material separation, which can cause changes in quality and unstable pumping. In particular, in hot summer conditions, the fluidity of the material decreases over time, and if the decrease in fluidity becomes too great, it can lead to blockage. Once a blockage occurs, it takes time to identify the blockage, clean up, and restore the pipe, which can easily hinder subsequent pumping and return work. If restoration work takes too long, the concrete may harden to the point where it cannot be removed, and the pipe itself may have to be abandoned. To prevent such a situation, it is necessary to monitor the pumpability of the concrete during pumping, predict the risk of blockage before it occurs, and take measures to restore good pumpability.

[0003] Regarding fluid pumping pipes that pump concrete to high altitudes (heights of about 200 m or more) or long distances, a technique is known for easily detecting the location of a blockage or leak in the event of a blockage or leak in the pipe (see, for example, Patent Document 1). In this technique, an optical fiber for measuring strain is helically laid on the outer periphery of the fluid pumping pipe, and a strain distribution measuring device measures the strain distribution in the length direction of the optical fiber. Then, the point where the strain distribution decreases is detected to identify the location of the blockage or leak in the fluid pumping pipe. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2023-005659 A Summary of the Invention [Problem to be solved by the invention]

[0005] There is a demand for a method to constantly evaluate and monitor the pumpability of concrete during pumping by monitoring the concrete during pumping. Pumpability here does not refer to the ease of pumping depending on the size of the pumping load of the concrete, but rather refers to the smooth pumping and continuity of a stable pumping state when pumping concrete to high altitudes (heights of about 200m or more) or long distances. Constant evaluation and monitoring includes judging whether the pumping state becomes unstable, including phenomena such as sudden blockages. An object of the present invention is to provide an evaluation system, a learning model creation device, an evaluation method, a learning model creation method, and a program capable of estimating the pumping state of concrete. [Means for solving the problem]

[0006] (1) One aspect of the present invention is an evaluation system for evaluating the pumping performance of concrete while it is being pumped through a transport pipe, the evaluation system comprising: a reception unit that receives pressure data detected by a pressure sensor disposed in a concrete pumping path of a transport pipe and detection time information at which the pressure sensor detected the pressure data; and an estimation unit that creates time series data of pressure data based on the pressure data and the detection time information received by the reception unit, and estimates a pumping state of the concrete while it is being pumped through the transport pipe based on a relationship between the created time series data of pressure data, the time series data of pressure data of the concrete pumping path, and information indicating the pumping state of the concrete. (2) One aspect of the present invention is the evaluation system described in (1) above, wherein the estimation unit estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of pressure data and the relationship between the stability of the pumping pressure of the time series data of pressure data of the concrete pumping route and the pumping state of the concrete. (3) One aspect of the present invention is the evaluation system described in (1) above, wherein the estimation unit estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of pressure data and the relationship between a stroke shape of the time series data of pressure data of the concrete pumping path and the pumping state of the concrete. (4) One aspect of the present invention is the evaluation system described in (1) above, wherein the estimation unit estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of pressure data and the relationship between the stroke end of the time series data of pressure data of the concrete pumping path and the pumping state of the concrete. (5) One aspect of the present invention is the evaluation system described in (1) above, in which the estimation unit has a trained model that has machine-learned the relationship between time-series data of pressure data in a concrete pumping route and information indicating the pumping state of the concrete, and estimates the pumping state of the concrete being pumped through the transport pipe based on the time-series data of pressure data and the trained model. (6) One aspect of the present invention is the evaluation system described in (1) above, wherein the pressure sensor is arranged at the base of the concrete pump. (7) One aspect of the present invention is the evaluation system described in (2) above, wherein the pressure sensor is positioned at the location where the concrete is poured and at the tip of the transport pipe. (8) One aspect of the present invention is the evaluation system described in any one of (5) to (7) above, wherein the estimation unit has a trained model that has machine-learned the relationship between the stability of pumping pressure of time-series data of pressure data of a concrete pumping route and information indicating the pumping state of the concrete. (9) One aspect of the present invention is the evaluation system described in any one of (5) to (7) above, wherein the estimation unit has a trained model that has machine-learned the relationship between the stroke shape of time-series data of pressure data in a concrete pumping path and information indicating the pumping state of the concrete. (10) One aspect of the present invention is the evaluation system described in any one of (5) to (7) above, wherein the estimation unit has a trained model that has machine-learned the relationship between the stroke end of the time-series data of pressure data in a concrete pumping path and information indicating the pumping state of the concrete.

[0007] (11) One aspect of the present invention is a learning model creation device including: a reception unit that receives a learning dataset including time series data of pressure data in a concrete pumping route as learning data and information indicating a pumping state of the concrete as teacher data; a processing unit that creates a learning model based on the learning dataset received by the reception unit by machine learning a relationship between the time series data of the pressure data in the concrete pumping route and the information indicating the pumping state of the concrete, with the time series data of the pressure data in the concrete pumping route as an explanatory variable and the information indicating the pumping state of the concrete as an objective variable; and an output unit that outputs the learning model created by the processing unit.

[0008] (12) One aspect of the present invention is an evaluation method that receives pressure data detected by a pressure sensor disposed in a concrete pumping path of a transport pipe and detection time information at which the pressure sensor detected the pressure data, creates time series data of pressure data based on the received pressure data and the detection time information, and estimates a pumping state of concrete being pumped through the transport pipe based on a relationship between the time series data of pressure data, the time series data of pressure data in the concrete pumping path, and information indicating the pumping state of the concrete.

[0009] (13) One aspect of the present invention is a learning model creation method executed by a computer, which receives a learning dataset including time series data of pressure data in a concrete pumping route as learning data and information indicating the concrete pumping state as teacher data, creates a learning model based on the received learning dataset by machine learning a relationship between the time series data of pressure data in a concrete pumping route and the information indicating the concrete pumping state, using the time series data of the pressure data in the concrete pumping route as an explanatory variable and the information indicating the concrete pumping state as a target variable, and outputs the created learning model.

[0010] (14) One aspect of the present invention is a program that causes a computer to accept pressure data detected by a pressure sensor disposed in a concrete pumping path of a transport pipe and detection time information at which the pressure sensor detected the pressure data, creates time series data of pressure data based on the accepted pressure data and the detection time information, and estimates the pumping state of concrete being pumped through the transport pipe based on a relationship between the time series data of pressure data, the time series data of pressure data in the concrete pumping path, and information indicating the pumping state of the concrete. (15) One aspect of the present invention is a program that causes a computer to accept a learning dataset including time series data of pressure data in a concrete pumping route as learning data and information indicating the concrete pumping state as teacher data, creates a learning model based on the accepted learning dataset by machine learning a relationship between the time series data of pressure data in a concrete pumping route and the information indicating the concrete pumping state, using the time series data of the pressure data in the concrete pumping route as an explanatory variable and the information indicating the concrete pumping state as a target variable, and outputs the created learning model. Effect of the Invention

[0011] According to the present invention, it is possible to obtain an effect of being able to estimate the pumping state of concrete. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 illustrates an example of an evaluation system 100 according to an embodiment of the present invention. [Figure 2A] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 2B] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 2C] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Diagram 3] FIG. 11 is a diagram illustrating an example of time-series data of pressure data. [Figure 4] 4 is a flowchart showing an example of the operation of the evaluation system 100 of the embodiment. [Diagram 5] FIG. 1 is a diagram illustrating an example of a learning model creation device 200 according to the present embodiment. [Figure 6] 4 is a flowchart showing an example of the operation of the learning model creation device 200 of the present embodiment. [Figure 7] FIG. 1 is a diagram illustrating an example of an evaluation system 100a according to a first modified example of the embodiment. [Figure 8A] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 8B] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 8C] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 8D] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 9] 13 is a flowchart showing an example of the operation of the evaluation system 100a of the first modified example of the embodiment. [Figure 10] 13 is a flowchart showing an example of the operation of the evaluation system 100a of the first modified example of the embodiment. [Figure 11]FIG. 13 is a diagram illustrating an example of an evaluation system 100b according to a second modified example of the embodiment. [Figure 12A] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 12B] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 12C] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. [Figure 12D] FIG. 11 is a diagram showing an example of time-series data of pressure data of a concrete pumping route. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Next, the evaluation system, learning model creation device, evaluation method, learning model creation method, and program of the present embodiment will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment. In addition, in all the drawings for explaining the embodiments, the same reference numerals are used for the parts having the same functions, and the repeated explanation is omitted. In addition, "based on XX" in this application means "based on at least XX," and includes cases where it is based on other elements in addition to XX. Furthermore, "based on XX" is not limited to cases where XX is used directly, but also includes cases where it is based on XX that has been calculated or processed. "XX" is any element (for example, any information).

[0014] (Embodiment) (Rating System) FIG. 1 is a diagram showing an example of an evaluation system 100 according to an embodiment of the present invention. The evaluation system 100 evaluates the pumping performance of concrete being pumped through a transport pipe. Pumping refers to moving a fluid or the like by the action of a pump pressure, and often refers in particular to the operation of pouring concrete (ready-mixed concrete) into a mold. For example, it includes pumping concrete to high altitudes (heights of about 200 m or more) and pumping concrete over long distances. A pressure sensor 50 is arranged in the pumping path of the concrete in the transport pipe. The evaluation system 100 receives pressure data detected by the pressure sensor 50 and detection time information at which the pressure sensor 50 detected the pressure data. The evaluation system 100 estimates the pumping state of the concrete being pumped through the transportation pipe, based on the received pressure data and detection time information.

[0015] The evaluation system 100 is realized by an apparatus such as a personal computer, a server, a smartphone, a tablet computer, an industrial computer, etc. The evaluation system 100 includes a communication unit 102, a reception unit 104, an estimation unit 106, an output unit 108, and a storage unit 110.

[0016] The communication unit 102 is realized by a communication module. The communication unit 102 communicates with an external communication device via the network NW. The communication unit 102 communicates using a wireless communication method such as wireless LAN, Bluetooth (registered trademark), or LTE (registered trademark). The communication unit 102 may also communicate using a communication method such as wired LAN. The communication unit 102 receives the pressure data and detection time information transmitted by the pressure sensor 50.

[0017] The storage unit 110 is realized by a hard disk drive (HDD), a flash memory, a random access memory (RAM), a read only memory (ROM), or the like. The receiving unit 104 acquires the pressure data and the detection time information from the communication unit 102. The receiving unit 104 accepts the acquired pressure data and detection time information.

[0018] The estimation unit 106 acquires the pressure data and the detection time information from the reception unit 104. The estimation unit 106 includes a trained model 107. The trained model 107 is a machine learning model of the relationship between the time series data of the pressure data of the concrete pumping path and information indicating the pumping state of the concrete. Specifically, the trained model 107 has machine-learned at least one of the following: the relationship between the stability of the pumping pressure in the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete; the relationship between the stroke shape in the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete; and the relationship between the stroke end in the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete. The information indicating the concrete pumping state includes a state where there is a margin for the pump capacity, a state where the margin for the pump capacity is running low, and a state where there is a possibility of closure.

[0019] 2A to 2C are diagrams showing an example of time-series data of pressure data in a concrete pumping route, in which the horizontal axis represents elapsed time [S] and the vertical axis represents pipe pressure [MPa]. When pumping starts, in the case of a piston-type concrete pump, the two pistons alternately push out repeatedly, so stable pressure waveform data can be obtained continuously. Therefore, if the fluidity of the concrete decreases and material separation occurs, causing smooth pumping, the stable waveform will be disturbed.

[0020] Figure 2A shows an example of time series data of pressure data in a state where there is a margin for the pump capacity. Figure 2A is an example of normal data. Under a pumping condition in which the pump is operated continuously, the pressure waveform is observed to have almost the same shape. Figure 2B shows an example of time series data of pressure data in a state where the pump capacity is running low. Figure 2B is an example of normal data. According to Figure 2B, slight differences in the magnitude of the pressure waveform are observed under continuous pumping conditions. Also, small surge pressures are observed. Figure 2C shows an example of time series data of pressure data in a blocked state. Figure 2B is an example of abnormal data. In a continuously operated pumping condition, even though the pump is pumping at a constant speed, the pressure continues to rise, and as shown by the circles, many of the waveforms at the stroke end are missing, causing the waveform to become distorted. Immediately after this situation, a blockage will occur. Let's return to Figure 1 and continue the explanation.

[0021] The estimation unit 106 creates time-series data of the pressure data based on the acquired multiple pressure data and detection time information. Fig. 3 is a diagram showing an example of time-series data of pressure data. In Fig. 3, the horizontal axis represents measurement time [S], and the vertical axis represents pipe pressure [MPa]. Returning to Fig. 1, the explanation will be continued. The estimation unit 106 estimates the pumping state of the concrete being pumped through the transport pipe based on the created time series data of the pressure data and the trained model 107. Specifically, the estimation unit 106 inputs the time series data of the pressure data to the trained model 107, and acquires information indicating the pumping state of the concrete output by the trained model 107. The estimation unit 106 outputs the acquired information indicating the pumping state of the concrete to the output unit 108.

[0022] The output unit 108 acquires the information indicating the pumping state of concrete output by the estimation unit 106. The output unit 108 outputs the acquired information indicating the pumping state of concrete. The output unit 108 may output the information indicating the pumping state of concrete by voice, or by displaying it on a display unit (not shown).

[0023] All or part of the reception unit 104, the estimation unit 106 and the output unit 108 are functional units (hereinafter referred to as software functional units) realized by a processor such as a CPU (Central Processing Unit) executing a program stored in the memory unit 110. In addition, all or part of the reception unit 104, the estimation unit 106, and the output unit 108 may be realized by hardware such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array), or may be realized by a combination of software functional units and hardware.

[0024] (Operation of the rating system 100) Fig. 4 is a flowchart showing an example of the operation of the evaluation system 100 of the embodiment. With reference to Fig. 4, the operation after the evaluation system 100 is installed before pumping and preparation for measurement such as attachment of the pressure sensor 50 is performed will be described. (Step S1-1) The communication unit 102 receives a plurality of pieces of pressure data and detection time information transmitted by the pressure sensor 50 . (Step S2-1) The receiving unit 104 acquires a plurality of pieces of pressure data and detection time information from the communication unit 102. The receiving unit 104 accepts the acquired plurality of pieces of pressure data and detection time information.

[0025] (Step S3-1) The estimation unit 106 acquires multiple pieces of pressure data and detection time information from the reception unit 104. The estimation unit 106 creates time-series data of pressure data based on the acquired multiple pieces of pressure data and detection time information. The estimation unit 106 estimates the pumping state of concrete being pumped through the transportation pipe based on the created time-series data of pressure data and the trained model 107. (Step S4-1) The output unit 108 acquires the information indicating the pumping state of the concrete output by the estimation unit 106. The output unit 108 outputs the acquired information indicating the pumping state of the concrete.

[0026] The creation of the trained model 107 will be described. The trained model 107 is created by a learning model creation device. The learning model creation device creates the trained model 107. Note that the evaluation system 100 may include the learning model creation device. In other words, the evaluation system 100 may create the trained model 107.

[0027] 5 is a diagram showing an example of the learning model creation device 200 of this embodiment. The learning model creation device 200 according to this embodiment is realized by a device such as a personal computer, a server, a smartphone, a tablet computer, or an industrial computer. The learning model creation device 200 trains a learning model (the model that serves as the basis for the learned model 107) using a learning dataset in which time-series data on the pressure data of the concrete pumping route is used as an input sample and information indicating the concrete pumping state is used as an output sample, thereby creating the learned model 107.

[0028] For example, the learning model creation device 200 uses algorithms such as CNN (Convolution Neural Network), RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), Random Forest, SVM (Support Vector Machine), and neural network to construct the learned model 107. An input sample is data input to an input layer when training a learning model. An output sample is data (teacher data) that is a correct answer to be compared with an output value output from an output layer when training a learning model. As an example, this embodiment focuses on CNN and uses CNN to construct the learned model 107.

[0029] The learning model creation device 200 includes an input unit 202, a receiving unit 204, a processing unit 206, an output unit 208, and a storage unit 210. The input unit 202 inputs information. As an example, the input unit 202 may have an operation unit such as a keyboard and a mouse. In this case, the input unit 202 inputs information according to an operation performed by a user on the operation unit. As another example, the input unit 202 may input information from an external device. The external device may be, for example, a portable storage medium. A learning dataset is input to the input unit 202.

[0030] The receiving unit 204 acquires a training dataset from the input unit 202 and accepts the acquired training dataset. The training dataset includes an input sample and an output sample, and the input sample and the output sample are paired. The training dataset is composed of a plurality of pairs.

[0031] The processing unit 206 inputs an input sample into the input layer of the learning model 207, calculates the error between the output value output from the output layer and the output sample (teacher data) corresponding to the input sample for each pair, changes the parameters of the learning model 207 (trains the learning model 207) so as to minimize the error, and creates the learned model 107. For example, the processing unit 206 trains the learning model 207 to learn at least one of the following by machine learning: the relationship between the stability of the pumping pressure in the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete; the relationship between the stroke shape of the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete; and the relationship between the stroke end of the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete.

[0032] The trained model 107 created as described above is received by the evaluation system 100 from the output unit 208 via a network (not shown) or a medium, and acquired by the estimation unit 106. When the learning model creation device 200 is included in the evaluation system 100, the estimation unit 106 acquires the trained model 107 from the learning model creation device 200.

[0033] All or part of the input unit 202, the receiving unit 204, the processing unit 206 and the output unit 208 are functional units (hereinafter referred to as software functional units) realized, for example, by a processor such as a CPU executing a program stored in the memory unit 210. In addition, all or part of the input unit 202, the reception unit 204, the processing unit 206, and the output unit 208 may be realized by hardware such as an LSI, an ASIC, or an FPGA, or may be realized by a combination of software functional units and hardware.

[0034] (Operation of the learning model creation device 200) Fig. 6 is a flowchart showing an example of the operation of the learning model creation device 200 of this embodiment. A process in which the learning model creation device 200 creates a trained model 107 by using a training data set prepared in advance will be described with reference to Fig. 6. (Step S1-2) The input unit 202 acquires a learning dataset. (Step S2-2) The receiving unit 204 acquires the training data set from the input unit 202 and accepts the acquired training data set.

[0035] (Step S3-2) The processing unit 206 acquires the learning dataset from the receiving unit 204. For every pair of an input sample and an output sample included in the learning dataset, the processing unit 206 inputs the input sample to the input layer of the learning model 207, calculates the error between the output value output from the output layer and the output sample (teacher data) corresponding to the input sample, changes the parameters of the learning model 207 (trains the learning model 207) so as to minimize the error, and creates the learned model 107. (Step S4-2) The output unit 208 acquires the trained model 107 from the processing unit 206. The output unit 208 outputs the acquired trained model 107.

[0036] In the above-described embodiment, as an example, the estimation unit 106 is provided with a trained model 107, time series data of pressure data is input to the trained model 107, and information indicating the pumping state of concrete output by the trained model 107 is obtained. However, this example is not limited to this. For example, the estimation unit 106 may estimate the pumping state of concrete being pumped through the transport pipe based on the relationship between the time series data of the pressure data, the time series data of the pressure data of the concrete pumping path, and the information indicating the pumping state of concrete. Specifically, the estimation unit 106 may derive a similarity between the time series data of the pressure data and the time series data of the pressure data of the concrete pumping path, and estimate the pumping state of concrete based on the derived similarity. For example, when the similarity between the time series data of the pressure data and the time series data of the pressure data of the concrete pumping path is equal to or greater than a threshold value, the estimation unit 106 acquires information indicating the pumping state of concrete associated with the time series data of the pressure data of the concrete pumping path.

[0037] The relationship between the time series data of the pressure data of the concrete pumping route and the information indicating the pumping state of the concrete may specifically be at least one of the relationship between the stability of the pumping pressure in the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete, the relationship between the stroke shape of the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete, and the relationship between the stroke end of the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete.

[0038] According to the evaluation system 100 of this embodiment, the evaluation system 100 evaluates the pumping performance of concrete being pumped through a transport pipe. The evaluation system 100 includes a reception unit 104 that receives a pressure sensor 50 detected by a pressure sensor arranged in a pumping path of the concrete in the transport pipe and detection time information at which the pressure sensor 50 detected the pressure data, and an estimation unit 106 that creates time-series data of the pressure data based on the pressure data and detection time information received by the reception unit 104, and estimates the pumping state of the concrete being pumped through the transport pipe based on the relationship between the created time-series data of the pressure data of the pumping path of the concrete and information indicating the pumping state of the concrete. By configuring in this manner, the evaluation system 100 can evaluate in real time the pumpability (stable, unstable state) of the concrete being pumped and signs of blockage, based on the pressure detected by the pressure sensor 50 arranged in the concrete pumping path of the transport pipe and the detection time information at which the pressure sensor 50 detected the pressure data.

[0039] In the evaluation system 100, the estimation unit 106 estimates the pumping state of concrete being pumped through the transport pipe based on the time series data of the pressure data and the relationship between the stability of the pumping pressure in the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete. By configuring in this manner, the evaluation system 100 can evaluate the pumpability of concrete during pumping (stable, unstable state) and signs of blockage in real time, based on the relationship between the stability of pumping pressure in the time-series data of the pressure data of the concrete pumping route and the pumping state of the concrete.

[0040] In the evaluation system 100, the estimation unit 106 estimates the pumping state of concrete being pumped through the transport pipe based on the time series data of the pressure data and the relationship between the stroke shape of the time series data of the pressure data of the concrete pumping path and the pumping state of the concrete. By configuring in this manner, the evaluation system 100 can evaluate in real time the pumpability (stable, unstable state) of the concrete during pumping and signs of blockage based on the relationship between the stroke shape of the time-series data of the pressure data of the concrete pumping path and the pumping state of the concrete.

[0041] In the evaluation system 100, the estimation unit 106 estimates the pumping state of concrete being pumped through the transport pipe based on the time series data of the pressure data and the relationship between the stroke end of the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete. By configuring in this manner, the evaluation system 100 can evaluate in real time the pumpability (stable, unstable state) of concrete during pumping and signs of blockage based on the relationship between the stroke end of the time-series data of the pressure data of the concrete pumping path and the pumping state of the concrete.

[0042] In the evaluation system 100, the estimation unit 106 has a trained model 107 that has machine-learned the relationship between the time series data of the pressure data of the concrete pumping route and information indicating the pumping state of the concrete, and estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of the pressure data and the trained model 107. By configuring in this manner, the evaluation system 100 can evaluate in real time the pumpability (stable, unstable state) of the concrete being pumped and signs of blockage, based on the pressure detected by the pressure sensor 50 arranged in the concrete pumping path of the transport pipe and the detection time information at which the pressure sensor 50 detected the pressure data.

[0043] In the evaluation system 100, the estimation unit 106 has a trained model 107 that has machine-learned the relationship between the stability of pumping pressure of the time-series data of the pressure data of the concrete pumping route and the pumping state of the concrete. By configuring in this manner, the evaluation system 100 can evaluate in real time the pumpability (stable, unstable state) of concrete during pumping and signs of blockage based on the trained model 107, which has machine-learned the relationship between the stability of pumping pressure of the time-series data of the pressure data of the concrete pumping route and the pumping state of the concrete.

[0044] In the evaluation system 100, the estimation unit 106 has a trained model 107 that has machine-learned the relationship between the stroke shape of the time-series data of the pressure data of the concrete pumping path and the pumping state of the concrete. By configuring in this manner, the evaluation system 100 can evaluate in real time the pumpability (stable, unstable state) of concrete during pumping and signs of blockage based on the trained model 107, which has machine-learned the relationship between the stroke shape of the time-series data of the pressure data in the concrete pumping path and the pumping state of the concrete.

[0045] In the evaluation system 100, the estimation unit 106 has a trained model 107 that has machine-learned the relationship between the stroke end of the time-series data of the pressure data of the concrete pumping path and the pumping state of the concrete. By configuring in this manner, the evaluation system 100 can evaluate in real time the pumpability (stable, unstable state) of concrete during pumping and signs of blockage based on the trained model 107, which has machine-learned the relationship between the stroke end of the time-series data of the pressure data in the concrete pumping path and the pumping state of the concrete.

[0046] According to the learning model creation device 200 of this embodiment, the learning model creation device 200 includes a reception unit 204 that receives a learning dataset including time series data of the pressure data of the concrete pumping route as learning data and information indicating the pumping state of the concrete as teacher data, a processing unit 206 that creates a learning model 207 by machine learning the relationship between the time series data of the pressure data of the concrete pumping route and the information indicating the pumping state of the concrete, based on the learning dataset received by the reception unit 204, with the time series data of the pressure data of the concrete pumping route as an explanatory variable and the information indicating the pumping state of the concrete as an objective variable, and an output unit 208 that outputs the learning model 207 created by the processing unit 206. By configuring it in this manner, a learning model can be created based on the learning dataset by machine learning the relationship between the time series data of the pressure data in the concrete pumping route and information indicating the concrete pumping status, using the time series data as the objective variable.

[0047] (First Modification of the Embodiment) (Rating System) 7 is a diagram showing an example of an evaluation system 100a according to the first modified example of the embodiment. The evaluation system 100a evaluates the pumping performance of concrete during pumping through a transport pipe. A pressure sensor 50a is disposed at a location (root) of the concrete pump path of the transport pipe close to the concrete pump body. The location of the concrete pump path of the transport pipe close to the concrete pump body is, for example, within about 10 m from the concrete pump body of the transport pipe. An example of the pressure sensor 50a is a flush diaphragm type pressure gauge. The evaluation system 100a receives pressure data detected by the pressure sensor 50a and detection time information when the pressure sensor 50a detected the pressure data. The evaluation system 100a estimates the pumping state of the concrete being pumped through the transportation pipe based on the received pressure data and detection time information.

[0048] The evaluation system 100a is realized by a device such as a personal computer, a server, a smartphone, a tablet computer, an industrial computer, etc. The evaluation system 100a includes a communication unit 102, a reception unit 104, an estimation unit 106a, an output unit 108, and a storage unit 110.

[0049] The communication unit 102 receives the pressure data and detection time information transmitted by the pressure sensor 50a. The estimation unit 106a acquires pressure data and detection time information from the reception unit 104. The estimation unit 106a includes a trained model 107a. The trained model 107a is a machine learning model of the relationship between time-series data of pressure data at a location close to the concrete pump body in the concrete pumping path and information indicating the pumping state of the concrete. Specifically, the trained model 107 has machine-learned at least one of the following: the relationship between the stability of the pumping pressure of the time series data of the pressure data at a point close to the concrete pump body in the concrete pumping path and the pumping state of the concrete; the relationship between the stroke shape of the time series data of the pressure data at a point close to the concrete pump body in the concrete pumping path and the pumping state of the concrete; and the relationship between the stroke end of the time series data of the pressure data at a point close to the concrete pump body in the concrete pumping path and the pumping state of the concrete. The information indicating the concrete pumping state includes a state in which there is a margin of pump capacity, a state in which the margin of pump capacity is running low, and a state in which there is a possibility of closure.

[0050] 8A to 8D are diagrams showing an example of time-series data of pressure data in a concrete pumping route, in which the horizontal axis represents measurement time [S] and the vertical axis represents pipe pressure [MPa]. When pumping starts, in the case of a piston-type concrete pump, the two pistons alternately push out repeatedly, so stable pressure waveform data can be obtained continuously. Therefore, if the fluidity of the concrete decreases and material separation occurs, causing smooth pumping, the stable waveform will be disturbed.

[0051] Figure 8A shows an example of time series data of pressure data in a state where there is a margin for the pump capacity. Figure 8A is an example of normal data. According to Figure 8A, under a pumping condition in which the pump is operated continuously, the pressure waveform is observed to have almost the same shape. In addition, the pressure increases, and a relatively small surge pressure is observed for the constant pressure value. If a visual evaluation reveals that surge pressure is present and the pressure levels are consistent, it can be determined that the pumping is extremely good.

[0052] FIG. 8B shows an example of time series data of pressure data in a state where the pump capacity is running low. FIG. 8B shows normal data. According to FIG. 8B, a slight difference in the magnitude of the pressure waveform is observed under a pumping condition in which the pump is continuously operated. Also, a small surge pressure is observed. Visual inspection revealed that surge pressure was present, but the shape was unstable and the pressure was not consistent. From these results, it can be determined that the pump is becoming unstable and is running out of capacity.

[0053] Figure 8C shows an example of time series data of pressure data in a state where there is a possibility of blockage. Figure 8C is between an example of normal data and abnormal data. According to Figure 8C, even though the pump is pumping at a constant speed during continuous operation, there are frequent points where the height changes along the way, and a pressure waveform with different heights for each stroke is observed. In addition, chipping is observed at the end of the stroke regardless of whether surge pressure is present. This is because the concrete is not being pushed out at a constant speed, and can be seen as a sign of blockage. Visual inspection revealed that the pressure waveform shape was unstable and the pressure height was continuously inconsistent. This result indicated that pumping performance was declining, and chipping at the stroke end was a sign of blockage.

[0054] FIG. 8D shows an example of time series data of pressure data in a blocked state. FIG. 8D is an example of abnormal data. According to FIG. 8D, even though the pump is pumping at a constant speed under continuous operation, the pressure continues to rise and the waveform at the stroke end is often missing. Immediately after the waveform becomes distorted, a blockage occurs. It is necessary to take measures before such a situation occurs.

[0055] The estimation unit 106a creates time-series data of the pressure data based on the acquired multiple pressure data and detection time information. The estimation unit 106a estimates the pumping state of the concrete being pumped through the transport pipe based on the created time series data of the pressure data and the trained model 107a. Specifically, the estimation unit 106a inputs the time series data of the pressure data to the trained model 107a and acquires information indicating the pumping state of the concrete output by the trained model 107a. The estimation unit 106a outputs the acquired information indicating the pumping state of the concrete to the output unit 108.

[0056] All or a part of the estimation unit 106a is a functional unit (hereinafter referred to as a software functional unit) that is realized by, for example, a processor such as a CPU executing a program stored in the storage unit 110. All or a part of the estimation unit 106a may be realized by hardware such as an LSI, an ASIC, or an FPGA, or may be realized by a combination of a software functional unit and hardware.

[0057] (Operation of evaluation system 100a) Fig. 9 is a flowchart showing an example of the operation of the evaluation system 100a of the embodiment modification 1. With reference to Fig. 9, the operation after the evaluation system 100a is installed before pumping and preparation for measurement such as attachment of the pressure sensor 50a is performed will be described. (Step S1-2) The communication unit 102 receives a plurality of pieces of pressure data and detection time information transmitted by the pressure sensor 50a. (Step S2-2) The receiving unit 104 acquires a plurality of pieces of pressure data and detection time information from the communication unit 102. The receiving unit 104 accepts the acquired plurality of pieces of pressure data and detection time information.

[0058] (Step S3-2) The estimation unit 106a acquires a plurality of pressure data and detection time information from the reception unit 104. The estimation unit 106a creates time-series data of pressure data based on the acquired plurality of pressure data and detection time information. The estimation unit 106a estimates the pumping state of concrete being pumped through the transport pipe based on the created time-series data of pressure data and the trained model 107a. (Step S4-2) The output unit 108 acquires the information indicating the pumping state of the concrete output by the estimation unit 106a. The output unit 108 outputs the acquired information indicating the pumping state of the concrete.

[0059] Fig. 10 is a flowchart showing an example of the operation of the evaluation system 100a of the first modified embodiment of the embodiment. Details of the process of step S3-2 in Fig. 9 will be described with reference to Fig. 10. (Step S3-21) The estimation unit 106a acquires a plurality of pressure data and detection time information from the reception unit 104. The estimation unit 106a creates time-series data of the pressure data based on the acquired plurality of pressure data and detection time information. (Step S3-22) The estimation unit 106a determines whether the pumping pressure is stable or not based on the created time-series data of the pressure data and the trained model 107a.

[0060] (Step S3-23) When it is determined that the pumping pressure is stable (step S3-22: YES), the estimation unit 106a determines whether or not there is a steady-state surge pressure. (Step S3-24) When it is determined that there is a steady-state surge pressure (step S3-23: YES), the estimation unit 106a determines that there is a margin for the pump capacity. In this case, the process of pumping concrete into the transportation pipe is continued.

[0061] (Step S3-25) If the estimation unit 106a determines that the pumping pressure is not stable (step S3-22: NO) or that there is no steady-state surge pressure (step S3-23: NO), it determines whether there are multiple disturbances in the stroke shape. (Step S3-26) When it is determined that there are not multiple disturbances in the shape of the stroke (step S3-25: NO), the estimation unit 106a determines whether there are multiple missing parts in the stroke.

[0062] (Step S3-27) When it is determined that there are not multiple stroke breaks (step S3-26: NO), the estimation unit 106a determines that the pump capacity is running low. In this case, the process of pumping concrete into the transport pipe is continued together with the evaluation by the evaluation system 100a.

[0063] (Step S3-28) If it is determined that there are multiple disturbances in the shape of the stroke (step S3-25: YES) or multiple missing parts of the stroke (step S3-26: YES), the estimation unit 106a determines that there is a possibility of closure. In this case, temporarily slow down the pumping speed and check the condition of the concrete being unloaded and discharged from the nozzle. If material separation or a decrease in fluidity is observed, unload the truck agitator with concrete in good condition. At this time, try to avoid stopping the pump as much as possible, and keep the concrete moving continuously even at a low speed to prevent a decrease in fluidity. While pumping at a low speed, gradually restore the pumping speed once it has been confirmed that the pressure waveform is stable, and continue construction.

[0064] The trained model 107a is created by the learning model creation device 200. That is, the learning model creation device 200 creates the trained model 107a. Note that the evaluation system 100a may include the learning model creation device 200. That is, the evaluation system 100a may create the trained model 107a.

[0065] In the above-mentioned embodiment, variant example 1 has been described in which the estimation unit 106a is provided with a trained model 107a, time series data of pressure data is input to the trained model 107a, and information indicating the pumping state of concrete output by the trained model 107a is obtained, but this example is not limited to this. For example, the estimation unit 106a may estimate the pumping state of concrete being pumped through the transport pipe based on the relationship between the time series data of the pressure data, the time series data of the pressure data of the concrete pumping path, and the information indicating the pumping state of concrete. Specifically, the estimation unit 106a may derive a similarity between the time series data of the pressure data and the time series data of the pressure data of the concrete pumping path, and estimate the pumping state of concrete based on the derived similarity. For example, when the similarity between the time series data of the pressure data and the time series data of the pressure data of the concrete pumping path is equal to or greater than a threshold value, the estimation unit 106a acquires information indicating the pumping state of concrete associated with the time series data of the pressure data of the concrete pumping path.

[0066] According to the evaluation system 100a according to the first modified example of the embodiment, the pressure sensor 50a is disposed at the base of the concrete pump in the evaluation system 100. The estimation unit 106a has a trained model 107a that has machine-learned the relationship between the time-series data of the pressure data at the base of the concrete pump and the pumping state of the concrete.

[0067] By configuring in this manner, the evaluation system 100a can evaluate in real time the pumpability (stable, unstable state) of the concrete being pumped and signs of blockage, based on the pressure data detected by the pressure sensor 50a located at a location near the concrete pump body in the concrete pumping path of the transport pipe and the detection time information when the pressure sensor 50a detected the pressure data.

[0068] (Modification 2 of the embodiment) (Rating System) 11 is a diagram showing an example of an evaluation system 100b according to the second modified example of the embodiment. The evaluation system 100b evaluates the pumping performance of concrete during pumping through a transport pipe. A pressure sensor 50a is disposed in a portion (root) of the pumping path of the concrete in the transport pipe close to the concrete pump body. A pressure sensor 50b is disposed in a portion (tip of the transport pipe) of the pumping path of the concrete in the transport pipe where the concrete is poured and where the piping does not move. For example, the location where the piping does not move in the concrete pouring path of the transport pipe (the tip of the transport pipe) is within about 10 m from the location where the concrete is poured in the transport pipe. In the case of a building structure, it is the pouring floor. One example of the pressure sensor 50b is a flush diaphragm type pressure gauge. The evaluation system 100b receives pressure data detected by the pressure sensor 50a and detection time information when the pressure sensor 50a detected the pressure data. The evaluation system 100b receives pressure data detected by the pressure sensor 50b and detection time information when the pressure sensor 50b detected the pressure data. The evaluation system 100b estimates the pumping state of the concrete being pumped through the transport pipe based on the received pressure data and detection time information.

[0069] The evaluation system 100b is realized by a device such as a personal computer, a server, a smartphone, a tablet computer, an industrial computer, etc. The evaluation system 100b includes a communication unit 102, a receiving unit 104, an estimation unit 106b, an output unit 108, and a storage unit 110.

[0070] The communication unit 102 receives the pressure data and detection time information transmitted by the pressure sensor 50a, and the pressure data and detection time information transmitted by the pressure sensor 50b. The estimation unit 106b acquires the pressure data and the detection time information from the reception unit 104. The estimation unit 106b includes a trained model 107b. The trained model 107b is machine-learned to learn the relationship between the time series data of the pressure data at a location close to the concrete pump body in the concrete pumping path and the time series data of the pressure data at a location in the concrete pumping path where the concrete is poured and where there is no movement of the piping, and the concrete pumping state.

[0071] Specifically, the trained model 107b has machine-learned at least one of the following: the relationship between the stability of the pumping pressure of the time series data of the pressure data at a location close to the concrete pump body in the concrete pumping path and the time series data of the pressure data at a location where the concrete is poured in the concrete pumping path where there is no movement of the piping and the pumping state of the concrete; the relationship between the stroke shape of the time series data of the pressure data at a location close to the concrete pump body in the concrete pumping path and the time series data of the pressure data at a location where the concrete is poured in the concrete pumping path where there is no movement of the piping and the pumping state of the concrete; and the relationship between the stroke end of the time series data of the pressure data at a location close to the concrete pump body in the concrete pumping path and the time series data of the pressure data at a location where the concrete is poured in the concrete pumping path where there is no movement of the piping and the pumping state of the concrete. The information indicating the concrete pumping state includes a state in which there is a margin of pump capacity, a state in which the margin of pump capacity is running low, and a state in which there is a possibility of closure.

[0072] When pumping starts, in the case of a piston-type concrete pump, the two pistons alternately push out repeatedly, so stable pressure waveform data can be obtained continuously. Therefore, if the fluidity of the concrete decreases and material separation occurs, causing smooth pumping, the stable waveform will be disturbed. For points on the pressure path close to the concrete pump body, the time series data of pressure data when there is a margin of error with respect to the pump capacity, the time series data of pressure data when there is no margin of error with respect to the pump capacity, the time series data of pressure data when there is a possibility of blockage, and the time series data of pressure data when blockage has occurred have been described above, so a description thereof will be omitted here. For points in the pumping route where concrete is poured and the piping does not move, we will explain time series data of pressure data when there is a margin for pump capacity, time series data of pressure data when there is no margin for pump capacity, time series data of pressure data when there is a possibility of blockage, and time series data of pressure data when the piping is blocked.

[0073] 12A to 12D are diagrams showing an example of time-series data of pressure data in a concrete pumping route, in which the horizontal axis represents measurement time [S] and the vertical axis represents pipe pressure [MPa]. When pumping starts, in the case of a piston-type concrete pump, the two pistons alternately push out repeatedly, so stable pressure waveform data can be obtained continuously. Therefore, if the fluidity of the concrete decreases and material separation occurs, causing smooth pumping, the stable waveform will be disturbed.

[0074] Figure 12A shows an example of time series data of pressure data in a state where there is a margin for the pump capacity. Figure 12A is an example of normal data. According to Figure 12A, under a pumping condition in which the pump is operated continuously, the pressure waveform is observed to have almost the same shape. Also, although the pressure itself is not large, a relatively large surge pressure is observed for a constant pressure value. If a visual evaluation reveals that surge pressure is present and the pressure levels are consistent, it can be determined that the pumping is extremely good.

[0075] FIG. 12B shows an example of time series data of pressure data in a state where the pump capacity is running low. FIG. 12B is an example of normal data. According to FIG. 12B, a slight difference in the magnitude of the pressure waveform is observed under a continuous pumping condition. Also, there are cases where the clear surge pressure that was observed during smooth pumping does not appear, and it does not appear continuously.

[0076] FIG. 12C shows an example of time series data of pressure data in a state where there is a possibility of blockage. FIG. 12C shows an example of data between normal data and abnormal data. According to FIG. 12C, even though the pump is pumped at a constant speed under continuous pumping operation, there are frequent points where the height changes along the way, and a pressure waveform with different heights for each stroke is observed. In addition, the clear surge pressure that was observed during smooth pumping is almost no longer observed. Visual inspection revealed that the pressure waveform shape was unstable and the pressure height was not consistent. If this condition continued, it could be determined that pumping performance was declining, and if chipping was observed at the stroke end, it could be determined that this was a sign of blockage.

[0077] Figure 12D shows an example of time series data of pressure data in the case of blockage. Figure 12D is an example of abnormal data. According to Figure 12D, even though the pump is pumping at a constant speed under continuous operation, the pressure continues to rise, and the stroke end waveform is often missing, resulting in blockage immediately after the waveform is disturbed. It is necessary to take measures before such a situation occurs.

[0078] The estimation unit 106b creates time-series data of the pressure data based on the acquired multiple pressure data and detection time information. The estimation unit 106b estimates the pumping state of the concrete being pumped through the transport pipe based on the created time series data of the pressure data and the trained model 107b. Specifically, the estimation unit 106b inputs the time series data of the pressure data to the trained model 107b and acquires information indicating the pumping state of the concrete output by the trained model 107b. The estimation unit 106b outputs the acquired information indicating the pumping state of the concrete to the output unit 108.

[0079] All or a part of the estimation unit 106b is a functional unit that is realized by a processor such as a CPU executing a program stored in the storage unit 110 (hereinafter, referred to as a software functional unit). All or part of the estimation unit 106b may be realized by hardware such as an LSI, an ASIC, or an FPGA, or may be realized by a combination of a software function unit and hardware.

[0080] The operation of the evaluation system 100b can be explained using FIGS. 9 and 10, and therefore a description thereof will be omitted. The trained model 107b is created by the learning model creation device 200. That is, the learning model creation device 200 creates the trained model 107b. Note that the evaluation system 100b may include the learning model creation device 200. That is, the evaluation system 100b may create the trained model 107b.

[0081] In the above-mentioned embodiment, variant example 2 has been described in which the estimation unit 106b is provided with a trained model 107b, time series data of pressure data is input to the trained model 107b, and information indicating the pumping state of concrete output by the trained model 107b is obtained, but this example is not limited to this. For example, the estimation unit 106b may estimate the pumping state of concrete being pumped through the transport pipe based on the relationship between the time series data of the pressure data, the time series data of the pressure data of the concrete pumping path, and the information indicating the pumping state of concrete. Specifically, the estimation unit 106b may derive a similarity between the time series data of the pressure data and the time series data of the pressure data of the concrete pumping path, and estimate the pumping state of concrete based on the derived similarity. For example, when the similarity between the time series data of the pressure data and the time series data of the pressure data of the concrete pumping path is equal to or greater than a threshold value, the estimation unit 106b acquires information indicating the pumping state of concrete associated with the time series data of the pressure data of the concrete pumping path.

[0082] According to the evaluation system 100b according to the second modification of the embodiment, the evaluation system 100b is the evaluation system 100a in which the pressure sensor 50b is disposed at the location where the concrete is poured and at the tip of the transport pipe. The estimation unit 106b has a trained model 107b that has machine-learned the relationship between the time-series data of the pressure data at the location where the concrete is poured and at the tip of the transport pipe and the pumping state of the concrete. By configuring in this manner, the evaluation system 100b can evaluate in real time the pumpability (stable, unstable state) of the concrete being pumped and signs of blockage based on the pressure data detected by the pressure sensor 50a positioned at the base of the concrete pump and the detection time information when the pressure sensor 50a detected the pressure data, and the pressure data detected by the pressure sensor 50b positioned at the tip of the transport pipe at the location where the concrete is poured and the detection time information when the pressure sensor 50b detected the pressure data.

[0083] Although the 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 design changes and the like within the scope of the present invention are also included. For example, a computer program for realizing the functions of each of the above-mentioned devices may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read into a computer system and executed. Note that the "computer system" referred to here may include hardware such as an OS and peripheral devices.

[0084] In addition, the term "computer-readable recording medium" refers to a storage device such as a flexible disk, an optical magnetic disk, a ROM, a writable non-volatile memory such as a flash memory, a portable medium such as a DVD, or a hard disk built into a computer system. Furthermore, the term "computer-readable recording medium" includes devices that retain a program for a certain period of time, such as volatile memory (e.g., DRAM (Dynamic Random Access Memory)) within a computer system that serves as a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line.

[0085] The program may be transmitted from a computer system in which the program is stored in a storage device or the like to another computer system via a transmission medium, or by transmission waves in the transmission medium. Here, the "transmission medium" that transmits the program refers to a medium that has a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may also be for realizing part of the above-mentioned functions. Furthermore, the above-mentioned functions may be realized in combination with a program already recorded in the computer system, that is, a so-called differential file (differential program). [Explanation of symbols]

[0086] 50, 50a, 50b...pressure sensor, 100, 100a, 100b...evaluation system, 102...communication unit, 104...reception unit, 106, 106a, 106b...estimation unit, 107, 107a, 107b...trained model, 108...output unit, 110...storage unit, 200...learning model creation device, 202...input unit, 204...reception unit, 206...processing unit, 207...learning model, 208...output unit, 210...storage unit

Claims

1. An evaluation system for evaluating pumping performance of concrete during pumping through a transport pipe, comprising: a reception unit that receives pressure data detected by a pressure sensor disposed in a concrete pumping path of the transport pipe and detection time information at which the pressure sensor detected the pressure data; an estimation unit that creates time series data of pressure data based on the pressure data and the detection time information received by the reception unit, and estimates a pumping state of the concrete being pumped through the transport pipe based on a relationship between the created time series data of pressure data, the time series data of pressure data of a concrete pumping route, and information indicating the pumping state of the concrete; An evaluation system comprising:

2. The evaluation system according to claim 1, wherein the estimation unit estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of the pressure data and a relationship between the stability of the pumping pressure of the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete.

3. The evaluation system according to claim 1 , wherein the estimation unit estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of the pressure data and a relationship between a stroke shape of the time series data of the pressure data of the concrete pumping route and the pumping state of the concrete.

4. The evaluation system according to claim 1 , wherein the estimation unit estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of pressure data and a relationship between a stroke end of the time series data of pressure data of a concrete pumping route and the pumping state of the concrete.

5. The evaluation system of claim 1, wherein the estimation unit has a trained model that has machine-learned the relationship between time series data of pressure data of a concrete pumping route and information indicating the pumping state of the concrete, and estimates the pumping state of the concrete being pumped through the transport pipe based on the time series data of pressure data and the trained model.

6. The evaluation system of claim 1 , wherein the pressure sensor is located at the root of a concrete pump.

7. The evaluation system according to claim 2 , wherein the pressure sensor is disposed at a location where the concrete is poured and at a tip of the transport pipe.

8. The evaluation system according to any one of claims 5 to 7, wherein the estimation unit has a trained model that machine-learns a relationship between stability of pumping pressure of time-series data of pressure data of a concrete pumping route and a pumping state of the concrete.

9. The evaluation system according to any one of claims 5 to 7, wherein the estimation unit has a trained model that machine-learns a relationship between a stroke shape of time-series data of pressure data of a concrete pumping path and a pumping state of the concrete.

10. The evaluation system according to any one of claims 5 to 7, wherein the estimation unit has a trained model that machine-learns a relationship between a stroke end of time-series data of pressure data of a concrete pumping path and a pumping state of the concrete.

11. A receiving unit that receives a learning data set including time series data of pressure data of a concrete pumping path as learning data and information indicating the concrete pumping state as teacher data; a processing unit that creates a learning model by machine learning a relationship between the time series data of the pressure data of the concrete pumping route and the information indicating the concrete pumping state, using the time series data of the pressure data of the concrete pumping route as an explanatory variable and the information indicating the concrete pumping state as an objective variable, based on the learning dataset received by the receiving unit; an output unit that outputs the learning model created by the processing unit; A learning model creation device comprising:

12. receiving pressure data detected by a pressure sensor disposed in a concrete pumping path of a transport pipe and detection time information at which the pressure sensor detected the pressure data; creating time-series data of the pressure data based on the received pressure data and the detection time information; An evaluation method for estimating the pumping state of concrete being pumped through the transport pipe based on the relationship between the time series data of pressure data, the time series data of pressure data of the concrete pumping route, and information indicating the pumping state of the concrete.

13. A learning data set is received, the learning data set including time series data of pressure data of a concrete pumping path as learning data and information indicating the concrete pumping state as teacher data; Based on the received learning dataset, a learning model is created by machine learning a relationship between the time series data of the pressure data of the concrete pumping route and the information indicating the concrete pumping state, with the time series data of the pressure data of the concrete pumping route being an explanatory variable and the information indicating the concrete pumping state being a target variable; A computer-implemented learning model creation method that outputs the created learning model.

14. On the computer, receiving pressure data detected by a pressure sensor disposed in a concrete pumping path of a transport pipe and detection time information at which the pressure sensor detected the pressure data; creating time-series data of the pressure data based on the received pressure data and the detection time information; A program that estimates the pumping state of concrete being pumped through the transport pipe based on the relationship between the time series data of pressure data, the time series data of pressure data of the concrete pumping route, and information indicating the pumping state of the concrete.

15. On the computer, Accepting a learning data set including time series data of pressure data of a concrete pumping route as learning data and information indicating the concrete pumping state as teacher data; Based on the received learning dataset, a learning model is created by machine learning a relationship between the time series data of the pressure data of the concrete pumping route and the information indicating the concrete pumping state, with the time series data of the pressure data of the concrete pumping route being used as an explanatory variable and the information indicating the concrete pumping state being used as an objective variable; A program for outputting the created learning model.

Images