Information processing device, determination method, and determination program

Abstract

An information processing device includes a first obtaining unit configured to obtain the temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool the raw material, from a temperature sensor installed on the outside of the pipe and a first determining unit configured to determine, based on the temperature of the pipe as obtained by the first obtaining unit, the presence or absence of the flow of the fluid through the pipe.

Description

INFORMATION PROCESSING DEVICE, DETERMINATION METHOD, AND DETERMINATION PROGRAM

[0001] The present invention relates to an information processing device, a determination method, and a determination program.

[0002] A manufacturing process can include a process for controlling the temperature of the raw material by heating or cooling the raw material using a fluid called a heat medium such as a heating medium or a cooling medium.

[0003] Japanese Laid-open Patent Publication No. 2017-80520

[0004] However, in the manufacturing process mentioned above, in spite of a stoppage in the manufacturing activity, such as in spite of the situation in which the raw material is not being inserted in a processing equipment such as a heating device or a cooling device; if the fluid representing the heating medium or the cooling medium is still supplied on a continuing basis, then it becomes difficult to hold down the wastage of energy.

[0005] It is an object of the present invention to support achieving reduction in the wastage of energy.

[0006] According to one aspect of embodiments, an information processing device includes: a first obtaining unit configured to obtain temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and a first determining unit configured to determine, based on the temperature of the pipe as obtained by the first obtaining unit, presence or absence of flow of the fluid through the pipe.

[0007] According to one aspect of embodiments, a determination method carried out by a computer, includes: obtaining temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and determining, based on the temperature of the pipe, presence or absence of flow of the fluid through the pipe.

[0008] According to one aspect of embodiments, a determination program causes a computer to execute a process including: obtaining temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and determining, based on the temperature of the pipe, presence or absence of flow of the fluid through the pipe.

[0009] According to an embodiment, it becomes possible to support achieving reduction in the wastage of energy.

[0010] Fig. 1 is a schematic diagram illustrating an example of a production facility.Fig. 2 is a block diagram illustrating an exemplary functional configuration of an information processing device.Fig. 3 is a schematic diagram illustrating an example of a machine learning model.Fig. 4 is a schematic diagram illustrating a determination case example regarding the energy usage condition.Fig. 5 is a flowchart for explaining the sequence of operations performed in a first determination operation.Fig. 6 is a flowchart for explaining the sequence of operations performed in a second determination operation.Fig. 7 is a flowchart for explaining the sequence of operations performed in a third determination operation.Fig. 8 is a diagram for explaining an exemplary hardware configuration.

[0011] An exemplary embodiment (hereinafter, referred to as an "embodiment") of an information processing device, a determination method, and a determination program according to the application concerned is described below in detail with reference to the accompanying drawings.　An embodiment merely represents examples and aspects; and the numerical values, the scope of the functions, and the usage situations are not limited by an embodiment.　Moreover, it is possible to adaptively combine embodiments without causing any contradictions in the operation details.

[0012] <Example of production facility> Fig. 1 is a schematic diagram illustrating an example of a production facility.　In Fig. 1 is illustrated a production facility 1 meant for implementing a manufacturing process in which the raw material, which is carried on a belt conveyer 3, is heated using a heating device 4.　With reference to Fig. 1, a usage situation is explained in which "steam" is used as the heating medium and a "heating device" is used as the processing equipment.　However, that is not the only possible case.　The heating device 4 is merely an example of a processing equipment meant for heating or cooling the raw material.　Thus, obviously, a usage situation in which a cooling device representing another example of the processing equipment performs cooling using a cooling medium, such as cooling water, is also included in the scope of the present embodiment.

[0013] As illustrated in Fig. 1, the production facility 1 includes the belt conveyer 3, a motor M, the heating device 4, and a pipe 5.　For example, the belt conveyer 3 is a device used for carrying the raw material, which is placed on a belt that rotates as a result of being driven by the motor M, in the direction of travel (the carrying direction) of the belt.　The heating device 4 is configured to heat the raw material, which is carried on the belt conveyer 3, using the steam supplied via the pipe 5.　The belt conveyer 3 is merely an example of a carrying mechanism, and the motor M is merely an example of a driving unit configured to drive the carrying mechanism.　Thus, the same functions can be implemented using some other mechanism or some other device.

[0014] <Aspect of problem> As explained earlier in the section of background art, in a manufacturing process, in spite of a stoppage in the manufacturing activity, such as in spite of the situation in which the raw material is not being inserted in a processing equipment such as a heating device or a cooling device, if the fluid representing the heating medium or the cooling medium is still supplied on a continuing basis, then it becomes difficult to hold down the wastage of energy.

[0015] <Aspect of approach for solving problem> As an approach for solving the problem mentioned above, it is possible to think of directly observing the physical quantity called the flow rate of the steam that is flowing through the pipe 5.　With that, the state in which the steam is flowing through the pipe 5 is monitored, that is, the energy usage condition is monitored.　Meanwhile, the approach for solving the problem as explained herein is distinguished from the conventional technology including already-disclosed Patent Literature, already-disclosed Non Patent Literature, and other commonly known technologies.

[0016] When the abovementioned problem solving approach is implemented, there is a disadvantage in the form of an increase in the man-hours required for installing a sensor for measuring the flow rate, such as a steam flow meter, and an increase in the cost attributed to the steam flow meter.　For example, a steam flow meter to be installed on the inside of the pipe 5 leads to an increase in the man-hours required for the installation.　Alternatively, a flow meter such as an ultrasonic flowmeter to be installed on the outer surface of the pipe 5 leads to an increase in the cost because of the expensive nature of the main body of the steam flow meter.

[0017] Such an increase in the man-hours and the cost is proportional to the number of pipes 5 that are to be monitored.　Hence, installing a steam flow meter only for the purpose of monitoring the energy usage condition is a difficult choice from the perspective of cost effectiveness.

[0018] In that regard, in the present embodiment, the technical significance lies in the fact of adapting a problem solving approach in which the energy usage condition is monitored using a mechanism that is not dependent on a steam flow meter configured to observe the flow rate as a physical quantity.

[0019] As such a problem solving approach, in the present embodiment, instead of obtaining the actual flow rate of the steam flowing through the pipe 5, a physical quantity that can be observed by a sensor installable on the outside of the pipe 5 is obtained as the physical quantity having a relationship with the flow rate.

[0020] Only as an example, an information processing device 10 obtains the surface temperature of the pipe 5 from a temperature sensor 5A that is installed on the outside of the pipe 5.　The temperature sensor 5A can be installed to make contact with the outer surface of the pipe 5.　However, the temperature sensor 5A need not always be installed in a contact state, and alternatively can be installed without making contact with the outer surface of the pipe 5.　For example, regarding the relationship between the surface temperature and the flow rate of the pipe 5, the surface temperature of the pipe 5 is higher when the steam is flowing therethrough as compared to the surface temperature when no steam is flowing therethrough.

[0021] As another example, the information processing device 10 obtains the vibration level of the pipe 5 from a vibration sensor 5B that is installed on the outside of the pipe 5.　For example, regarding the relationship between the vibration level and the flow rate of the pipe 5, the vibration level of the pipe 5 is higher when the steam is flowing therethrough as compared to the vibration level when no steam is flowing therethrough.

[0022] Based on the surface temperature and the vibration level of the pipe 5, the information processing device 10 determines whether or not the steam is flowing through the pipe 5.　Regarding such a task of two-class sorting, even if the flow rate of the steam flowing through the pipe 5 is not provided as the input, the task can be implemented in a sufficient and highly accurate manner as long as a physical quantity related to the flow rate is input.　Herein, although the explanation is given about using the surface temperature of the pipe 5 as well as using the vibration level of the pipe 5, it is possible to use at least either the surface temperature of the pipe 5 or the vibration level of the pipe 5.

[0023] As a result, if it is detected that the steam is flowing through the pipe 5 in spite of the fact that the raw material is not being carried on the belt conveyer 3, the information processing device 10 can output a warning.

[0024] As an example of the output destination, it is possible to think of an information processing terminal (hereinafter, referred to as a "task participant terminal") 20 that is used by a task participant 2 such as a worker at the production facility 1 or a field supervisor who oversees the production facility 1.

[0025] For example, the task participant terminal 20 can be implemented using a mobile handheld device such as a smartphone or a tablet terminal, or using a wearable terminal, or using a personal computer such as a desktop computer or a notebook computer.　In that case, the information processing device 10 can display an icon or a message as a warning about wasting the energy or a warning about prompting an operation for stopping the supply of the steam into the pipe 5.　However, a warning is not limited to a display output, and can be in the form of a sound output or a print output.

[0026] Other than outputting a warning to the task participant terminal 20, the information processing device 10 can output a warning to a device included in the production facility 1 illustrated in Fig. 1 or to an output unit such as a display unit, a sound output unit, or a print output unit of a device not illustrated in Fig. 1.

[0027] Meanwhile, the explanation herein is given about outputting a warning to prompt an operation for stopping the supply of the steam into the pipe 5.　However, alternatively, the control for stopping the supply of the steam into the pipe 5 can be performed automatically.

[0028] As explained above, the information processing device 10 according to the present embodiment can support achieving reduction in the wastage of energy during a stoppage in the production activity.　Moreover, since monitoring of the energy usage condition is achieved using a mechanism not dependent on a steam flow meter configured to observe the flow rate as a physical quantity, it also becomes possible reduce the man-hours and the cost related to the installation of sensors.　The impact of achieving such reduction in the man-hours and the cost is boosted accompanying an increase in the number of pipes 5 that are to be monitored.

[0029] <Configuration of information processing device 10> Given below is the explanation about an exemplary functional configuration of the information processing device 10 according to the present embodiment.　Fig. 2 is a block diagram illustrating an exemplary functional configuration of the information processing device 10.　As illustrated in Fig. 2, the information processing device 10 is communicably connected to a wattmeter 3A, the temperature sensor 5A, the vibration sensor 5B, and the task participant terminal 20.

[0030] The group of sensors including the wattmeter 3A, the temperature sensor 5A, and the vibration sensor 5B can be configured to perform communication with the information processing device 10 according to an industrial wireless communication standard.　The task participant terminal 20 and the information processing device 10 can be connected to each other by an arbitrary type of network such as the Internet or a LAN (Local Area Network).

[0031] Between the information processing device 10 and the group of sensors, it is not always necessary to have two-way communication enabled.　Alternatively, serial communication from the group of sensors to the information processing device 10 can be implemented.　Moreover, the communication implemented between the information processing device 10 and the group of sensors is not limited to be of a specific communication standard such as an industrial communication standard, and there is no restriction as far as wired communication or wireless communication is concerned.

[0032] In Fig. 2 are illustrated schematic blocks related to the monitoring function provided in the information processing device 10 for monitoring the energy usage condition.　As illustrated in Fig. 2, the information processing device 10 includes a communication control unit 11, a memory unit 13, and a control unit 15.　Meanwhile, Fig. 2 is used merely for selectively illustrating the functional units related to the monitoring function, and other functional units not illustrated in Fig. 2 can also be included in the information processing device 10.

[0033] The communication control unit 11 is a functional unit configured to control the communication between the group of sensors, which includes the wattmeter 3A, the temperature sensor 5A, and the vibration sensor 5B, and other devices such as the task participant terminal 20.　Only as an example, the communication control unit 11 can be implemented using a network interface card.　As an aspect, the communication control unit 11 can receive information from the wattmeter 3A about the amount of power consumed in the motor M, can receive information from the temperature sensor 5A about the surface temperature of the pipe 5, and can receive information from the vibration sensor 5B about the vibration level of the pipe 5.　As another aspect, the communication control unit 11 can output, to the task participant terminal 20, a warning about the wastage of energy during a stoppage in the production activity.

[0034] The memory unit 13 is a functional unit configured to store a variety of data.　Only as an example, the memory unit 13 is implemented using an internal storage, or an external storage, or an auxiliary storage of the information processing device 10.　The memory unit 13 is configured to store, for example, a machine learning model 13M.　Regarding the machine learning model 13M, the explanation is given later along with the explanation of the situations in which the machine learning model 13M is referred to, is generated, and is implemented.

[0035] The control unit 15 is a functional unit configured to perform the overall control of the information processing device 10.　For example, the control unit 15 can be implemented using a hardware processor.　As illustrated in Fig. 2, the control unit 15 includes a first obtaining unit 15A, a second obtaining unit 15B, a third obtaining unit 15C, a first determining unit 15D, a second determining unit 15E, and a third determining unit 15F.　Meanwhile, the control unit 15 can be alternatively implemented using a hard-wired logic.

[0036] The first obtaining unit 15A is a processing unit configured to obtain the surface temperature of the pipe 5 from the temperature sensor 5A.　The second obtaining unit 15B is a processing unit configured to obtain the vibration level of the pipe 5 from the vibration sensor 5B.　The third obtaining unit 15C is a processing unit configured to obtain, from the wattmeter 3A, the amount of power consumed by the motor M.　Herein, the power consumption of the motor M is given as an example of the index value of the electric power used in determining the energy input condition.　However, the power consumption of the motor M is only exemplary.　Alternatively, the third obtaining unit 15C can obtain some other index value of the electric power related to the motor M or related to a driving unit similar to the motor M.　For example, the third obtaining unit 15C can obtain the index value of some other item such as the electric current or the voltage consumed by the motor M or the reactance of the motor M.　Meanwhile, instead of obtaining the index value of the electric power consumed by the motor M, the third obtaining unit 15C can obtain the index value of the electric power supplied to the motor M.

[0037] The first obtaining unit 15A, the second obtaining unit 15B, and the third obtaining unit 15C can obtain, in real time, the sensor values output by the sensors such as the temperature sensor 5A, the vibration sensor 5B, and the wattmeter 3A; or can obtain time-series data of the sensor values over a specific duration.

[0038] The first determining unit 15D is a processing unit that, based on the surface temperature of the pipe 5 as obtained by the first obtaining unit 15A and the vibration level of the pipe 5 as obtained by the second obtaining unit 15B, determines whether or not the steam is flowing through the pipe 5, that is, determines the energy usage condition.

[0039] Only as an example, such determination of the energy usage condition can be performed using the machine learning model 13B that executes a class sorting task in which the surface temperature and the vibration level of the pipe 5 are treated as the input and the confidence factor such as "presence" or "absence" of the steam flow is output on a class-by-class basis.

[0040] In the following explanation, only as an example, the abovementioned machine learning model is implemented using a neural network.　However, alternatively, the abovementioned machine learning model can be implemented using a support vector machine or a gradient boosting decision tree.

[0041] Fig. 3 is a schematic diagram illustrating an example of the machine learning model 13M.　As illustrated in Fig. 3, for the determination of the energy usage condition, the already-trained machine learning model 13M is used that is obtained as a result of training an untrained machine learning model 13m according to a machine learning algorithm such as deep learning.

[0042] In the training of the untrained machine learning model 13m, it is possible to use a dataset 13TR that includes training data in which the surface temperature and the vibration level of the pipe 5 are held in a corresponding manner to a correct solution label of "presence" or "absence" of the steam flow.

[0043] For example, in Fig. 3, as an example of the dataset 13TR, three sets of training data are selectively illustrated in which time-series data of the temperature and time-series data of the vibration level are held in a corresponding manner to each correct solution label from among two types of correct solution labels indicating "presence" and "absence" of the steam flow.

[0044] Herein, the time-series data of the temperature and the time-series data of the vibration level can be data strings in which the data values of the temperature and the data values of the vibration level during a specific duration, such as five seconds, are respectively arranged in chronological order.　Among the two data strings, it is possible to match the timing of the initial data point, the timing of the last data point, the number of data points, and the timing of each data.　However, it is also possible to have mismatches within the allowable range.　That is, the sampling period of the temperature sensor 5A need not always be identical to the sampling period of the vibration sensor 5B.　Moreover, the size of the window for clipping the time-series data from the signals output from each sensor can be different among the sensors, and the temporal positions at which windows are applied to the signals can be different among the sensors.

[0045] In the following explanation, the time-series data about the surface temperature of the pipe 5 is sometimes referred to as "temperature data", and the time-series data about the vibration level of the pipe 5 is sometimes referred to as "vibration data".

[0046] For example, in the training phase, the temperature data and the vibration data are treated as the explanatory variables of the untrained machine learning model 13m; the correct solution label is treated as the objective variable of the untrained machine learning model 13m; and an arbitrary machine learning algorithm, such as deep learning, is used for training the untrained machine learning model 13m.　Then, the already-trained machine learning model 13M, which is obtained as a result of the training, is registered in the memory unit 13.　For example, the memory unit 13 is used to store the following: hyper parameters related to the laminar structure such as the neurons or the synapse of the input layer constituting the already-trained machine learning model 13M; and parameters related to the objective function, such as the weights and the biases of various layers.

[0047] In the inference phase, the temperature data obtained by the first obtaining unit 15A and the vibration data obtained by the second obtaining unit 15B is input to the already-trained machine learning model 13M.　With the temperature data and the vibration data input thereto, the already-trained machine learning model 13M outputs the class-by-class confidence factor such as "presence" or "absence" of the steam flow.　In the example illustrated in Fig. 3, the confidence factor of "90%" is output regarding the "presence" of the steam flow.　Moreover, the confidence factor of "10%" is output regarding the "absence" of the steam flow.　Meanwhile, the explanation herein is given about an example in which the time-series data of the temperature and the time-series data of the vibration level are input to the already-trained machine learning model 13M.　However, it is not always necessary that the time-series data of the temperature and the time-series data of the vibration level are input.　Alternatively, for example, feature quantities, such as the average, the dispersion, and the median value, obtained as a result of performing feature extraction with respect to the time-series data of the temperature and the time-series data of the vibration level can be input to the already-trained machine learning model 13M.

[0048] Based on the confidence factor indicating "presence" of the steam flow, the first determining unit 15D determines the energy usage condition in the production facility 1.　For example, the first determining unit 15D determines whether or not the confidence factor indicating "presence" of the steam flow is equal to or greater than a threshold value Th1.　If the confidence factor indicating "presence" of the steam flow is equal to or greater than the threshold value Th1, then the first determining unit 15D outputs, to the third determining unit 15F, "presence" of the steam flow as the determination result about the energy usage condition.　On the other hand, if the confidence factor indicating "presence" of the steam flow is neither equal to nor greater than the threshold value Th1, then the first determining unit 15D outputs, to the third determining unit 15F, "absence" of the steam flow as the determination result about the energy usage condition.

[0049] Fig. 4 is a schematic diagram illustrating a determination case example regarding the energy usage condition.　In Fig. 4 is illustrated a graph G1 in which the waveform of the signal output by the temperature sensor 5A is plotted using a solid line, and the waveform of the signal output by the vibration sensor 5B is plotted using a dashed line.　In the graph G1, the vertical axis represents the sensor value of the surface temperature or the vibration level of the pipe 5, and the horizontal axis represents the time.　In the graph G1, windows W1 to W4 are schematically illustrated within which the temperature data or the vibration data is clipped at four timings t1 to t4, respectively, from the signal of each sensor.　Moreover, in Fig. 4 is illustrated a graph G2 for plotting the time-series data of the confidence factor indicating "presence" of the steam flow as output by the already-trained machine learning model 13M.　In the graph G2, the vertical axis represents the confidence factor and the horizontal axis represents time.　In the graph G2, the lower limit value of the confidence factor indicating "presence" of the steam flow is illustrated as the threshold value Th1.　Meanwhile, in Fig. 4, the windows W1 to W4 are assumed to have the window width, that is, the duration equal to 5 seconds.

[0050] For example, at the timing t1, the window W1 is set in such a way that the timing t1 arrives at the end (the right end) of the window W1.　In that case, from the signal output by each of the temperature sensor 5A and the vibration sensor 5B, the waveform of the portion corresponding to the window width of the window W1, that is, the data string of sensor values within the duration from the timing arriving five seconds earlier than the timing t1 to the timing t1 is clipped as the temperature data and the vibration data, respectively.　Then, the already-trained machine learning model 13M, to which the clipped temperature data and the clipped vibration data has been input, outputs the confidence factor indicating "presence" of the steam flow.　Herein, since the confidence factor indicating "presence" of the steam flow is equal to or greater than the threshold value Th1, it is determined that "presence" of the steam flow represents the energy usage condition.

[0051] Moreover, at the timings t2, t3, and t4 too, in an identical manner to the timing t1, the confidence factor indicating "presence" of the steam flow is obtained.　For example, at the timing t2, since the confidence factor indicating "presence" of the steam flow is equal to or greater than the threshold value Th1, it is determined that "presence" of the steam flow represents the energy usage condition.　On the other hand, at the timing t3, since the confidence factor indicating "presence" of the steam flow is smaller than the threshold value Th1, it is determined that "absence" of the steam flow represents the energy usage condition.　Similarly, at the timing t4 too, since the confidence factor indicating "presence" of the steam flow is smaller than the threshold value Th1, it is determined that "absence" of the steam flow represents the energy usage condition.

[0052] Herein, the operations at the four timings t1 to t4 are explained in a sectional manner.　As far as the entire period of time is concerned, the following explanation holds true.　As illustrated in the lower part of the graph G2, in the section in which a hatched band is illustrated as a solid line, "presence" of the steam flow is determined.　On the other hand, in the section in which a hatched band is illustrated as a dashed line, "absence" of the steam flow is determined.

[0053] Meanwhile, the explanation herein is given about an example in which two types of data, namely, the temperature data and the vibration data are input to the already-trained machine learning model 13M.　However, it is not always necessary that each of the two types of data is input to the already-trained machine learning model 13M.　That is, with only either the temperature data or the vibration data input thereto, the already-trained machine learning model 13M can be configured to output the class either indicating "presence" of the steam flow or indicating "absence" of the steam flow.

[0054] Meanwhile, the explanation herein is given about determining the energy usage condition using the already-trained machine learning model 13M.　However, the determination of the energy usage condition can be performed also without using a machine learning model.　For example, the first determining unit 15D can determine the energy usage condition based on the slope or the intercept of an approximation straight line that is obtained as a result of implementing regression analysis, such as linear regression, with respect to the temperature data or the vibration data.　More particularly, when the slope has the "positive" sign, it is more likely that the steam is flowing through the pipe 5.　Moreover, when the slope is within a predetermined range from zero, it is more likely that either the flow of the steam is stable or the state of no steam flow is going on.　At that time, if the intercept is a positive value equal to or greater than a certain value, then it is more likely that the flow of the steam is stable.　On the other hand, when the intercept is within a predetermined range from zero, it is more likely that the state of no steam flow is going on.　In this way, when the slope has the "positive" sign and is not within a predetermined range from zero, "presence" of the steam flow can be determined.　Moreover, when the slope is within a predetermined range from zero and when the intercept is a positive value equal to or greater than a threshold value, "presence" of the steam flow can be determined.

[0055] The second determining unit 15E is a processing unit that, based on the power consumption of the motor M as obtained by the third obtaining unit 15C, determines whether or not the raw material is being carried on the belt conveyer 3, that is, determines the insertion condition of the raw material.

[0056] Only as an example, the second determining unit 15E determines whether or not the power consumption of the motor M is equal to or greater than a threshold value Th2.　If the power consumption of the motor M is equal to or greater than the threshold value Th2, then the second determining unit 15E outputs, to the third determining unit 15F, "being inserted" regarding the raw material as the determination result about the raw material insertion condition.　On the other hand, if the power consumption of the motor M is neither equal to nor greater than the threshold value Th2, then the second determining unit 15E outputs, to the third determining unit 15F, "not being inserted" regarding the raw material as the determination result about the raw material insertion condition.

[0057] The third determining unit 15F is a processing unit that, based on the determination result about the energy usage condition as obtained by the first determining unit 15D and based on the determination result about the raw material insertion condition as obtained by the second determining unit 15E, determines whether or not there is any wastage of energy during a stoppage in the manufacturing activity.

[0058] Only as an example, the third determining unit 15F determines whether or not the energy usage condition indicates "presence" of the steam flow and whether or not the raw material insertion condition indicates that the raw material is "not being inserted".　If the energy usage condition indicates "presence" of the steam flow and if the raw material insertion condition indicates that the raw material is "not being inserted", then it can be recognized that there is wastage of energy during a stoppage in the production activity.　In that case, the third determining unit 15F outputs, to arbitrary output destinations including the task participant terminal 20, a warning about the wastage of energy or a warning to prompt an operation for stopping the supply of the steam into the pipe 5.

[0059] <Flow of operations> Given below is the explanation of a flow of the operations performed in the information processing device 10 according to the present embodiment.　The following explanation is given about (1) First determination operation, (2) Second determination operation, and (3) Third determination operation.

[0060] (1) First determination operation Fig. 5 is a flowchart for explaining the sequence of operations performed in the first determination operation.　Herein, only as an example, the first determination operation either can be performed when sensor values are obtained from the temperature sensor 5A and the vibration sensor 5B, or can be performed at regular intervals.

[0061] As illustrated in Fig. 5, the first obtaining unit 15A obtains the surface temperature of the pipe 5 from the temperature sensor 5A, and the second obtaining unit 15B obtains the vibration level of the pipe 5 from the vibration sensor 5B (Step S101 and Step S102).

[0062] Then, to the already-trained machine learning model 13M, the first determining unit 15D inputs the temperature data that is for a predetermined period of time in the past and that contains the surface temperature obtained at Step S101 as well as inputs the vibration data that is for a predetermined period of time in the past and that contains the vibration level obtained at Step S101 (Step S103).　Moreover, the first determining unit 15D determines whether or not the confidence factor indicating "presence" of the steam flow as output by the already-trained machine learning model 13M is equal to or greater than the threshold value Th1 (Step S104).

[0063] If the confidence factor indicating "presence" of the steam flow is equal to or greater than the threshold value Th1 (Yes at Step S104), then the first determining unit 15D outputs, to the third determining unit 15F, "presence" of the steam flow as the determination result about the energy usage condition (Step S105).　That marks the end of the operations.

[0064] On the other hand, if the confidence factor indicating "presence" of the steam flow is neither equal to nor greater than the threshold value Th1 (No at Step S104), then the first determining unit 15D outputs, to the third determining unit 15F, "absence" of the steam flow as the determination result about the energy usage condition (Step S106).　That marks the end of the operations.

[0065] Meanwhile, regarding the operations performed at Step S101 and Step S102 illustrated in the flowchart in Fig. 5, the sequence is not limited to the sequence illustrated in Fig. 5.　That is, those operations can be performed in random order or can be performed simultaneously.

[0066] (2) Second determination operation Fig. 6 is a flowchart for explaining the sequence of operations performed in the second determination operation.　Herein, only as an example, the second determination operation either can be performed when the power consumption of the motor M is obtained from the wattmeter 3A or can be performed at regular intervals.

[0067] As illustrated in Fig. 6, the third obtaining unit 15C obtains the power consumption of the motor M from the wattmeter 3A (Step S301).　Then, the second determining unit 15E determines whether or not the power consumption of the motor M as obtained at Step S301 is equal to or greater than the threshold value Th2 (Step S302).

[0068] If the power consumption of the motor M is equal to or greater than the threshold value Th2 (Yes at Step S302), then the second determining unit 15E outputs, to the third determining unit 15F, "being inserted" regarding the raw material as the determination result about the raw material insertion condition (Step S303).　That marks the end of the operations.

[0069] On the other hand, if the power consumption of the motor M is neither equal to nor greater than the threshold value Th2 (No at Step S302), then the second determining unit 15E outputs, to the third determining unit 15F, "not being inserted" regarding the raw material as the determination result about the raw material insertion condition (Step S304).　That marks the end of the operations.

[0070] (3) Third determination operation Fig. 7 is a flowchart for explaining the sequence of operations performed in the third determination operation.　Herein, only as an example, the third determination operation either can be performed when the first determining unit 15D has determined the energy usage condition and when the second determining unit 15E has determined the raw material insertion condition, or can be performed at regular intervals.

[0071] As illustrated in Fig. 7, the third determining unit 15F determines whether or not the determination result about the energy usage condition as obtained by the first determining unit 15D indicates "presence" of the steam flow (Step S501).　If the determination result indicates "absence" of the steam flow (No at Step S501), then that marks the end of the operations.

[0072] On the other hand, if the determination result indicates "presence" of the steam flow (Yes at Step S501), then the third determining unit 15F determines whether or not the determination result about the raw material insertion condition as obtained by the second determining unit 15E indicates that the raw material is "not being inserted" (Step S502).　If the determination result indicates that the raw material is "being inserted" (Yes at Step S502), then that marks the end of the operations.

[0073] On the other hand, if the determination result indicates that the raw material is "not being inserted" (No at Step S502), then it can be recognized that there is wastage of energy during a stoppage in the production activity.　In that case, the third determining unit 15F outputs, to arbitrary output destinations including the task participant terminal 20, a warning about the wastage of energy or a warning to prompt an operation for stopping the supply of the steam into the pipe 5 (Step S503).　That marks the end of the operations.

[0074] Meanwhile, regarding the operations performed at Step S501 and Step S502 illustrated in the flowchart in Fig. 7, the sequence is not limited to the sequence illustrated in Fig. 7.　That is, those operations can be performed in random order or can be performed simultaneously.

[0075] <Aspect of effects> As explained above, based on the surface temperature and the vibration level of the pipe 5, the information processing device 10 according to the present embodiment determines whether or not the steam is flowing through the pipe 5.　Hence, for example, if it is detected that the steam is flowing through the pipe 5 in spite of the fact that the raw material is not being carried on the belt conveyer 3, then the information processing device 10 can output a warning.　Thus, the information processing device 10 according to the present embodiment becomes able to support achieving reduction in the wastage of energy during a stoppage in the manufacturing activity.　Moreover, since the energy usage condition is monitored using a mechanism that is not dependent on a steam flow meter configured to observe the flow rate as a physical quantity, it also becomes possible to reduce the man-hours and the cost related to the installation of sensors.　The impact of achieving such reduction in the man-hours and the cost is boosted accompanying an increase in the number of pipes 5 that are to be monitored.

[0076] <Numerical values> The items explained in the embodiment, such as the number of sensors and specific examples of the training method and the inference method for the machine learning model 13M are only exemplary and can be modified.　Moreover, the flowcharts explained in the embodiment can have a different sequence of operations without causing any contradictions.

[0077] <System> The processing procedures, the control procedures, specific names, various data, and information including parameters described in the embodiment or illustrated in the drawings can be changed as required unless otherwise specified.　For example, from among the first determining unit 15D, the second determining unit 15E, and the third determining unit 15F; one or more functional units can be configured using separate devices.

[0078] The constituent elements of the device illustrated in the drawings are merely conceptual, and need not be physically configured as illustrated.　The constituent elements, as a whole or in part, can be separated or integrated either functionally or physically based on various types of loads or use conditions.　Meanwhile, each configuration can be a physical configuration.

[0079] The process functions implemented in the device are entirely or partially implemented by a CPU (Central Processing Unit) or by programs that are analyzed and executed by a CPU, or are implemented as hardware by wired logic.

[0080] <Hardware> Given below is the explanation of an exemplary hardware configuration of a computer according to the embodiment.　Fig. 8 is a diagram for explaining an exemplary hardware configuration.　As illustrated in Fig. 8, the information processing device 10 includes a communication device 10a, a hard disk drive (HDD) 10b, a memory 10c, and a processor 10d.　The constituent elements illustrated in Fig. 8 are connected to each other by a bus.

[0081] The communication device 10a is a network interface card that performs communication with other servers.　The HDD 10b is used to store programs and databases meant for implementing the functions illustrated in Fig. 2.

[0082] The processor 10d reads a program, which is written for executing the operations identical to the processing units illustrated in Fig. 2, from the HDD 100b and loads it in the memory 100c.　As a result, a process is run that is meant for implementing the functions explained with reference to Fig. 2.　For example, the process implements functions identical to the processing units included in the information processing device 10.　More particularly, the processor 10d reads, from the HDD 10b, a program that is equipped with identical functions to the first obtaining unit 15A, the second obtaining unit 15B, the third obtaining unit 15C, the first determining unit 15D, the second determining unit 15E, and the third determining unit 15F.　Then, the processor 10d executes a process that implements the operations identical to the first obtaining unit 15A, the second obtaining unit 15B, the third obtaining unit 15C, the first determining unit 15D, the second determining unit 15E, and the third determining unit 15F.

[0083] In this way, the information processing device 10 operates as an information processing device that reads and executes a program and implements a factor analysis method.　Alternatively, the information processing device 10 can read the abovementioned program from a recording medium using a medium reading device, execute the read program, and implement the functions identical to the embodiment described above.　Meanwhile, the program according to the other embodiment is not limited to be executed by the information processing device 10.　For example, even when some other computer or a server executes the program or when such devices execute the program in cooperation, the present invention can still be applied in an identical manner.

[0084] Still alternatively, the abovementioned program can be distributed via a network such as the Internet.　Still alternatively, the abovementioned program can be recorded in an arbitrary recording medium, so that a computer can read the program from the recording medium and execute it.　For example, the recording medium can be implemented using a hard disk, a flexible disk (FD), a CD-ROM, an MO (Magneto-Optical disk), or a DVD (Digital Versatile Disc).

[0085] <Miscellaneous> Given below is the explanation of some combinations of the technical features disclosed herein.

[0086] (1) 　　　An information processing device comprising: 　　　a first obtaining unit configured to obtain temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and 　　　a first determining unit configured to determine, based on the temperature of the pipe as obtained by the first obtaining unit, presence or absence of flow of the fluid through the pipe.

[0087] (2) 　　　The information processing device according to (1), further comprising a second obtaining unit configured to obtain vibration level of the pipe from a vibration sensor installed on outside of the pipe, wherein 　　　the first determining unit determines, based on the temperature of the pipe as obtained by the first obtaining unit and based on the vibration level of the pipe as obtained by the second obtaining unit, presence or absence of flow of the fluid through the pipe.

[0088] (3) 　　　The information processing device according to (2), further comprising: 　　　a third obtaining unit configured to obtain information about electric power consumed by a driving unit, which is configured to drive a carrying mechanism configured to carry the raw material to the processing equipment, or to obtain index value of electric power supplied to the driving unit; 　　　a second determining unit configured to determine, based on the index value of electric power as obtained by the third obtaining unit, whether the raw material is being inserted or is not being inserted; and 　　　a third determining unit configured to determine whether or not determination result obtained by the first determining unit indicates presence of flow of the fluid and to determine whether or not determination result obtained by the second determining unit indicates that the raw material is not being inserted.

[0089] (4) 　　　The information processing device according to (3), wherein, when presence of flow of the fluid is indicated and when the raw material is not being inserted, the third determining unit outputs a warning.

[0090] (5) 　　　The information processing device according to any one of (1) to (4), wherein the temperature sensor is installed to make contact with outer surface of the pipe.

[0091] (6) 　　　The information processing device according to any one of (1) to (5), wherein, based on output that is obtained when time-series data of temperature of the pipe as obtained by the first obtaining unit is input to a machine learning model configured to perform class sorting about presence or absence of flow of the fluid, the first determining unit determines presence or absence of flow of the fluid through the pipe.

[0092] (7) 　　　The information processing device according to (6), wherein, based on whether or not confidence factor of class indicating presence of flow of the fluid as output by the machine learning model is equal to or greater than a threshold value, the first determining unit determines presence or absence of flow of the fluid through the pipe.

[0093] (8) 　　　The information processing device according to (6) or (7), wherein the machine learning model is trained as a result of execution of machine learning in which time-series data of temperature of the pipe is treated as explanatory variable of the machine learning model and a label indicating presence or absence of flow of the fluid is treated as objective variable of the machine learning model.

[0094] (9) 　　　A determination method carried out by a computer, comprising: 　　　obtaining temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and 　　　determining, based on the temperature of the pipe, presence or absence of flow of the fluid through the pipe.

[0095] (10) 　　　A determination program that causes a computer to execute a process comprising: 　　　obtaining temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and 　　　determining, based on the temperature of the pipe, presence or absence of flow of the fluid through the pipe.

[0096] 1　　　production facility 2　　　task participant 3　　　belt conveyer 3A　　wattmeter 4　　　heating device 5　　　pipe 5A　　temperature sensor 5B　　vibration sensor 10　　information processing device 11　　communication control unit 13　　memory unit 13M　machine learning model 15　　control unit 15A　first obtaining unit 15B　second obtaining unit 15C　third obtaining unit 15D　first determining unit 15E　second determining unit 15F　third determining unit 20　　task participant terminal

Claims

1. An information processing device comprising: 　　　a first obtaining unit configured to obtain temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and 　　　a first determining unit configured to determine, based on the temperature of the pipe as obtained by the first obtaining unit, presence or absence of flow of the fluid through the pipe.

2. The information processing device according to claim 1, further comprising a second obtaining unit configured to obtain vibration level of the pipe from a vibration sensor installed on outside of the pipe, wherein 　　　the first determining unit determines, based on the temperature of the pipe as obtained by the first obtaining unit and based on the vibration level of the pipe as obtained by the second obtaining unit, presence or absence of flow of the fluid through the pipe.

3. The information processing device according to claim 2, further comprising: 　　　a third obtaining unit configured to obtain information about electric power consumed by a driving unit, which is configured to drive a carrying mechanism configured to carry the raw material to the processing equipment, or to obtain index value of electric power supplied to the driving unit; 　　　a second determining unit configured to determine, based on the index value of electric power as obtained by the third obtaining unit, whether the raw material is being inserted or is not being inserted; and 　　　a third determining unit configured to determine whether or not determination result obtained by the first determining unit indicates presence of flow of the fluid and to determine whether or not determination result obtained by the second determining unit indicates that the raw material is not being inserted.

4. The information processing device according to claim 3, wherein, when presence of flow of the fluid is indicated and when the raw material is not being inserted, the third determining unit outputs a warning.

5. The information processing device according to claim 1, wherein the temperature sensor is installed to make contact with outer surface of the pipe.

6. The information processing device according to any one of claims 1 to 5, wherein, based on output that is obtained when time-series data of temperature of the pipe as obtained by the first obtaining unit is input to a machine learning model configured to perform class sorting about presence or absence of flow of the fluid, the first determining unit determines presence or absence of flow of the fluid through the pipe.

7. The information processing device according to claim 6, wherein, based on whether or not confidence factor of class indicating presence of flow of the fluid as output by the machine learning model is equal to or greater than a threshold value, the first determining unit determines presence or absence of flow of the fluid through the pipe.

8. The information processing device according to claim 6, wherein the machine learning model is trained as a result of execution of machine learning in which time-series data of temperature of the pipe is treated as explanatory variable of the machine learning model and a label indicating presence or absence of flow of the fluid is treated as objective variable of the machine learning model.

9. A determination method carried out by a computer, comprising: 　　　obtaining temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and 　　　determining, based on the temperature of the pipe, presence or absence of flow of the fluid through the pipe.

10. A determination program that causes a computer to execute a process comprising: 　　　obtaining temperature of a pipe, which is used to supply a fluid representing either a heating medium or a cooling medium to a processing equipment configured to heat or cool raw material, from a temperature sensor installed on outside of the pipe; and 　　　determining, based on the temperature of the pipe, presence or absence of flow of the fluid through the pipe.

Images