INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING SYSTEM, AND PROGRAM, FOR IoT

Abstract

To reduce power consumption in an information processing device and the like for IoT.SOLUTION: An information processing device includes a first detection unit that detects vibration or acceleration of a rotating device that operates irregularly, a second detection unit that detects operation or stop of the rotating device, a determination unit that generates determination data by determining a state of the rotating device that has signs of abnormality based on detection data that indicates detection results by the first detection unit, a communication unit that transmits or receives the detection data or the determination data, and a control unit that has two or more control modes and switches the control mode depending on the second detection unit to control the first detection unit and the communication unit.SELECTED DRAWING: Figure 7

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing system, and a program for IoT (Internet of Things).

Background Art

[0002] Efforts are being made to promote the Sustainable Development Goals (the 2030 Agenda for Sustainable Development, adopted at the United Nations Summit on September 25, 2015 (Heisei 27), hereinafter referred to as "SDGs"). Specifically, in order to spread inexpensive and reliable energy, technologies for achieving power saving are required.

[0003] Conventionally, in a wireless communication system composed of a master device that communicates with a slave device driven by a battery, first, the master device transmits a clear including sleep time information to the slave device. Then, the slave device sets the sleep time based on the sleep time information and measures the sleep time. Also, when setting the sleep time, the power supply of the main circuit is turned off, and when the sleep time has elapsed, the power supply of the main circuit is turned on. By synchronizing the sleep cycles between the master device and the slave device in this way, two-way communication is enabled, and a technique for improving the battery life of the slave device is known (for example, Patent Document 1, etc.).

[0004] Also, in a sensor unit, a vibration sensor is fixed directly to a monitoring object in a non-contact state with a case. With such a configuration, only the vibration of the monitoring object can be surely detected without being affected by the vibration of the case. In this way, a technique is known that enables easy attachment to the monitoring object and detection of a dominant frequency higher than 1 kHz (for example, Patent Document 2, etc.).

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] By using the conventional technology, it is possible to monitor the states of devices that operate constantly or periodically with low power consumption. However, in the case of a device that operates irregularly and has a short operation time per operation, it is difficult to monitor the state of the device during operation. Although it is possible to monitor the state during operation by shortening the sleep period, there is a problem that the power consumption may increase significantly.

[0007] An object of the present invention is to reduce the power consumption in an information processing device for IoT or the like.

Means for Solving the Problems

[0008] The information processing device includes a first detection unit that detects the vibration or acceleration of a rotating device that operates irregularly, and a second detection unit that detects whether the rotating device is operating or stopped, and a determination unit that determines a state including a sign of abnormality of the rotating device based on detection data indicating the detection result by the first detection unit and generates determination data, and a communication unit that transmits or receives the detection data or the determination data, and a control unit that has two or more control modes and switches the control mode according to the second detection unit to control the first detection unit and the communication unit and includes.

Effects of the Invention

[0009] According to the present invention, the power consumption in an information processing device for IoT or the like can be reduced.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments will be described with reference to the drawings. In the following description, elements related to the invention will be described in detail, and other elements will be omitted. The same elements are denoted by the same reference numerals.

[0012] Also, in the following examples, it is assumed that the rotating device is the pump device 1 operating in a factory or the like. However, the rotating device is not limited to the pump device 1. For example, the rotating device may be a prime mover, a driven machine, or a speed increaser / decreaser, etc. Specifically, the prime mover is a motor or a turbine, etc. The driven machine is, for example, a pump or an agitator, etc. The speed increaser / decreaser is, for example, a gear, etc. Note that the rotating device is not limited to a device composed only of a prime mover, a driven machine, or a speed increaser / decreaser, and may be a device including these devices.

[0013] [Configuration Example of Information Processing System] FIG. 1 is a diagram showing a configuration example of an information processing system. The information processing system 10 is composed of a sensor unit 2, a server 3, an output device 4, etc. However, the information processing system 10 may have devices other than these.

[0014] In the information processing system 10, the sensor unit 2, the server 3, the output device 4, etc. transmit and receive data by wireless communication. However, the information processing system 10 may have a wired network.

[0015] The sensor unit 2 is an example of an information processing device installed for devices to be maintained and monitored, such as the pump device 1. For example, when a plurality of pump devices 1 are the targets, one sensor unit 2 is installed for each pump device 1. However, one sensor unit 2 may be commonly used for a plurality of pump devices 1.

[0016] Hereinafter, for the purpose of determining the state of the pump device 1, the data generated by the sensor unit 2 through measurement is referred to as "sensor data". Note that the sensor data is added with identification information for identifying which sensor unit 2 generated it, and it is identified which pump device 1 the data indicates the state of.

[0017] In the information processing system 10, it is desirable that devices installed in the factory, such as the sensor unit 2 and the output device 4, are waterproof, dustproof, explosion-proof, and temperature-adaptive. In particular, since the environment where the rotating device is installed handles moisture, the waterproofing standards are often stricter than those of general machine tools, etc. Also, the rotating device can often be judged by vibration compared with other machine tools, etc. Therefore, the information processing system 10 makes a judgment targeting the rotating device (including peripheral devices provided or connected to the pump device 1, etc.) based on vibration.

[0018] Also, the information processing system 10 may have an information processing device other than the server 3. For example, there may be an edge server, etc.

[0019] [Example of Hardware Configuration of Sensor Unit] FIG. 2 is a diagram showing an example of the hardware configuration of the sensor unit. For example, the sensor unit 2 has a hardware configuration including a first acceleration sensor 2H1, a control circuit board 2H2, a communication module 2H3, a battery 2H4, a second acceleration sensor 2H5, and the like. Note that the sensor unit 2 may further include other devices inside or outside thereof.

[0020] The first acceleration sensor 2H1 measures the vibration or acceleration generated in the pump device 1 and generates sensor data. Specifically, the first acceleration sensor 2H1 is specified to have a measurement range of frequencies up to about 10 kilohertz (kHz) for at least one axis in the gravitational direction. However, the measurement conditions of the first acceleration sensor 2H1 may vary depending on an actuator or the like provided in the pump device 1.

[0021] Hereinafter, the data indicating the detection result by the first acceleration sensor 2H1 is referred to as "detection data". For example, when the first acceleration sensor 2H1 measures vibration, the detection data is data showing the vibration generated in the pump device 1 in chronological order (that is, the result of measuring what kind of vibration occurred over a certain period of time). Note that what kind of type the detection data indicates, how the data is generated, and each parameter and the like are set in advance.

[0022] The control circuit board 2H2 is a control device that controls the hardware included in the sensor unit 2 and the output device 4 connected to the sensor unit 2, and an arithmetic device that performs processes such as data processing. The control circuit board 2H2 may be provided with a storage device and the control circuit board 2H2 may be able to store data.

[0023] The communication module 2H3 communicates with an external device such as a server 3. For example, the communication module 2H3 is Wi-Fi (registered trademark), Bluetooth (registered trademark), LPWA (Low Power Wide Area, low power wide area network), or a combination thereof. Note that the communication standard used by the communication module 2H3 may be other than these.

[0024] Also, it may communicate with devices other than the server 3. For example, in a configuration with an edge server or the like, if there is an edge server between the communication module 2H3 and sensors such as the first acceleration sensor 2H1 and the server 3, communication may be performed with the edge server or the like.

[0025] Specifically, it is desirable that the edge server is configured to receive data from the sensor and perform preprocessing and data compression processing. That is, it is desirable that the data after preprocessing and data compression processing is sent to the server 3. In this way, the presence of an edge server that performs preprocessing and data compression processing can reduce the network load. Performing preprocessing before performing the communication process of transmitting data to the server 3 by the communication module 2H3 can reduce the processing performed on the server 3 side and reduce the load on the server 3. Furthermore, by deleting unnecessary data (for example, data that becomes overhead) or performing compression processing that is reversible compression, irreversible compression, or a combination of these so that the data capacity becomes smaller, the data capacity of the data transmitted to the server 3 by the communication module 2H3 can be reduced, and traffic can be reduced.

[0026] The battery 2H4 serves as the power source for the hardware included in the sensor unit 2 and the output device 4. Note that the battery 2H4 switches the power supply state based on the control of the control circuit board 2H2 or the like. Specifically, the battery 2H4, under the control of the control circuit board 2H2, places the hardware that consumes the power supplied by the battery 2H4 (hereinafter referred to as the "power consumption part". Note that the power consumption part may include an external device that supplies power to the output device 4 or the like) in a low-consumption state.

[0027] Hereinafter, through control, the power consumption is reduced to a "power-saving state" (which may also be referred to as "sleep mode" or "stop state", etc.). The power-saving state only needs to have a lower power consumption than the "start state" described later. Specifically, the power-saving state is the so-called "sleep state", which has a lower power consumption than the "start state", but does not completely consume no power. Instead, it is a state in which power is supplied so that it can function and switch to the "start state" through control.

[0028] On the other hand, the power-saving state may also be a state in which power supply is stopped. Hereinafter, an example in which the power-saving state is the "sleep state" will be described.

[0029] The state in which the sleep state is released and sufficient power is supplied from the battery 2H4 so that the functions provided in each device function to the extent defined in the specifications is called the "start state" (which may also be referred to as "normal mode" or "operating", etc.).

[0030] Hereinafter, the control to bring the first acceleration sensor 2H1 or the communication module 2H3 into the start state (a part may also be in the start state) is called the "first control mode". On the other hand, the control to put the first acceleration sensor 2H1 or the communication module 2H3 into the sleep state (a part may also be in the sleep state) and other power-saving states is called the "second control mode". Note that the detailed settings such as which devices the first control mode and the second control mode target and how much power is saved are set in advance (when there are three or more control modes, other control modes are set in the same way).

[0031] Therefore, due to the first control mode, the first acceleration sensor 2H1, the communication module 2H3, etc. enter the start state, and the state transitions to a state in which functions such as data acquisition or communication are exerted.

[0032] On the other hand, in the second control mode, when the first acceleration sensor 2H1, the communication module 2H3, etc. enter the sleep state, the power consumption of the first acceleration sensor 2H1, the communication module 2H3, etc., and the power consumption of the device that realizes the sensing or communication function, which is large, can be reduced. Therefore, during the time period when processing such as sensing and communication is not performed, the sensor unit 2 is placed in a state where its functions are partially stopped by the second control mode.

[0033] Note that there may be three or more control modes. For example, in addition to the above-described first control mode and second control mode, there may be a third control mode. For example, the third control mode is performed when the control circuit board 2H2 is in the power-saving state and the second acceleration sensor 2H5 is in the active state.

[0034] In addition, even when the first acceleration sensor 2H1 and the communication module 2H3 are in a state such as sleep, that is, "inactive", there may be a state in which the wake-up signal receiving unit in the wake-up processing unit and the control unit is "active" and startup can be performed.

[0035] Note that there may be a plurality of batteries 2H4. For example, in a configuration with a plurality of batteries 2H4, the plurality of batteries 2H4 may be used separately in the sleep state and the startup state, or may be divided in terms of supply destinations.

[0036] The second acceleration sensor 2H5 detects whether the pump device 1 is operating or stopped. Specifically, the second acceleration sensor 2H5 detects, for example, the operation of the pump device 1 by acceleration. Hereinafter, when the second acceleration sensor 2H5 detects the operation of the pump device 1, the signal output by the second acceleration sensor 2H5 is referred to as "Trigger".

[0037] Note that the second acceleration sensor 2H5 does not have to be a sensor. That is, the second acceleration sensor 2H5 may be a control circuit or the like that issues a trigger in response to the operation of the pump device 1.

[0038] For example, the second acceleration sensor 2H5 is a sensor that detects a magnetic field generated when the pump device 1 starts operating. Alternatively, the second acceleration sensor 2H5 is a sensor that detects radio waves generated when the pump device 1 starts operating. Specifically, like passive RFID (Radio Frequency Identification), when the second acceleration sensor 2H5 senses the occurrence of radio waves, it outputs a trigger by the energy of the radio waves.

[0039] Similarly, the second acceleration sensor 2H5 may detect that the pump device 1 has finished operating and entered a standby state. For example, when the second acceleration sensor 2H5 can no longer detect vibrations from the pump device 1, it detects that the pump device 1 has entered a standby state.

[0040] The second acceleration sensor 2H5 detects that the pump device 1 has started operating or detects signs of starting to operate. Therefore, the second acceleration sensor 2H5 mainly detects initial vibrations and the like that occur when the pump device 1 starts operating.

[0041] Note that a configuration in which an amplification circuit, a filter circuit, etc. are provided after the second acceleration sensor 2H5 may also be used.

[0042] The second acceleration sensor 2H5 is preferably controlled to switch between a mode of detecting the operation or standby of the pump device 1 (hereinafter referred to as the "first detection mode") and a mode of not detecting the operation or standby of the pump device 1 (hereinafter referred to as the "second detection mode").

[0043] For example, the first detection mode and the second detection mode are switched at a predetermined period. The period is set to be sufficiently shorter than the interval at which the pump device 1 operates. Since the pump device 1 operates irregularly, for example, after it operates, it operates next after an interval of several days to several months. In this case, the second acceleration sensor 2H5 is switched to the first detection mode, for example, at a period of about several minutes or several hours. If the period is sufficiently short like this, even when the pump device 1 operates irregularly, the start of operation can be detected without omission.

[0044] Also, the second acceleration sensor 2H5 does not necessarily need to periodically switch between the first detection mode and the second detection mode. The pump device 1 operates irregularly. Therefore, the pump device 1 operates once and then operates next under non-periodic conditions. Thus, the second acceleration sensor 2H5 may switch between the first detection mode and the second detection mode under conditions that match the conditions of the pump device 1.

[0045] Specifically, the second acceleration sensor 2H5 learns or sets conditions such as the conditions under which the pump device 1 operates, and based on those conditions, switches between the first detection mode and the second detection mode aiming at the time when the pump device 1 is likely to operate.

[0046] Alternatively, the second acceleration sensor 2H5 may be a sensor that consumes less power than the communication module 2H3 and the first acceleration sensor 2H1, and the second acceleration sensor 2H5 may be in a state of constantly performing detection.

[0047] The output device 4 is, for example, a lighting device such as an LED (Light Emitting Diode). Additionally, the output device 4 may also be a display or an audio output device, etc.

[0048] The output device 4 outputs the determination result step by step. For example, the output device 4 emits light of multiple colors. Therefore, the output device 4 outputs, for example, "normal", "warning", or "abnormal" in different colors if the determination result is one of these, and makes it an output indicating a level. Note that the step output may be realized by the blinking method, the message display on the display, or the brightness of the light, etc. Thus, it is desirable that the output device 4 has a configuration for step output using three or more levels such as "normal", "warning", or "abnormal".

[0049] Note that the hardware configuration is not limited to the above configuration. For example, the output device 4 and the control circuit board 2H2 may be integrated. Also, the control circuit board 2H2 may be two or more devices, etc.

[0050] [Example of Server Hardware Configuration] FIG. 3 is a diagram showing an example of the hardware configuration of a server. For example, the server 3 has a hardware configuration including a Central Processing Unit (hereinafter referred to as "CPU3H1"), a storage device 3H2, an input device 3H3, an output device 3H4, and a communication device 3H5, etc.

[0051] The CPU3H1 is an arithmetic device and a control device that performs arithmetic operations and control based on a program.

[0052] The storage device 3H2 is a main storage device such as a memory. Note that the storage device 3H2 may have an auxiliary storage device such as a hard disk.

[0053] The input device 3H3 is a device that inputs the operations of an operator. For example, the input device 3H3 is a connector that receives an input operation signal, a keyboard, a mouse, or the like.

[0054] The output device 3H4 is a device that outputs the processing result to the operator. For example, the output device 3H4 is a connector that outputs an output signal, a display, or the like.

[0055] The communication device 3H5 is a device that communicates with an external device by wire or wirelessly. For example, the communication device 3H5 is a connector or the like.

[0056] Note that the server 3 may further include hardware other than those described above. For example, the server 3 may have a hardware configuration that includes an arithmetic device, a control device, a storage device, an input device, an output device, a communication device, or an auxiliary device externally or internally. Also, the server 3 may be composed of a plurality of devices.

[0057] The server 3 may be a so-called "cloud server" or the like. Also, the server 3 may be a public server or a private server.

[0058] [Example of Control Timing] FIG. 4 is a diagram showing an example of control timing. Hereinafter, the state signal SG1, the trigger signal SG2, and the mode signal SG3 indicating the state of the pump device 1 will be described with a timing chart. In this example, the information processing system 10 performs control to switch the mode based on the trigger signal SG2, and switches between the sleep state and the startup state.

[0059] If the state signal SG1 is "High", the pump device 1 is in an operating state. On the other hand, if the state signal SG1 is "Low", the pump device 1 is in a non-operating state.

[0060] The trigger signal SG2 is a signal that becomes active (in this example, the trigger signal SG2 is a high-active signal) when the pump device 1 is detected to be operating. For example, when it is detected that the pump device 1 starts operating, such as at the first timing T1 or the second timing T2, the trigger signal SG2 is output.

[0061] The mode signal SG3 indicates the mode. Specifically, when the mode signal SG3 is "High", it indicates that the power consumption part is in the startup state. On the other hand, when the mode signal SG3 is "Low", it indicates that the power consumption part is in the sleep state.

[0062] When the trigger signal SG2 is output, mode control is performed to switch the power consumption part from the sleep state to the startup state, such as the third timing T3 and the fourth timing T4.

[0063] Note that the power consumption part performs mode control to switch from the startup state to the sleep state when, for example, processing for determining the state of the pump device 1, that is, when determination data is generated.

[0064] For example, in the sleep state, the power supplies of all of the first acceleration sensor 2H1, the control circuit board 2H2, the communication module 2H3, and the output device 4 are set to "OFF". However, the communication module 2H3 or a part of the device is set to the startup state so as to accept wake-up from the outside.

[0065] In addition, the condition for transitioning from the startup state to the sleep state (hereinafter referred to as "sleep trigger") is preferably, for example, when it is determined that communication for transmitting data to the server 3 has been completed. Specifically, the sleep trigger is realized by, for example, a determination result that the first acceleration sensor 2H1 has stopped detecting vibration or the like, or a completion notification that the server 3 has normally received data.

[0066] When receiving the sleep trigger issued in the above cases, the sensor unit 2 transitions to the sleep state. Therefore, with the above sleep trigger, it is possible to prevent, for example, transitioning to the sleep state during communication and incomplete data transmission and reception.

[0067] Furthermore, the sleep state may be a state with low power consumption in addition to turning off the power supply. That is, the sleep state may be to stop a part of the device or switch to a mode with low power consumption. For example, in the sleep state, control such as not supplying power from the power supply to the target element or the like, stopping the power supply (also referred to as a state where the power supply has dropped), or setting the operation clock in the operation to a low frequency is performed to reduce the power consumption.

[0068] The power supply system may have its power supply controlled separately for a plurality of systems. For example, assume that the power supply system has two systems. Specifically, among the two power supply systems, the first power supply system is the power supply system for an arithmetic device such as a microcomputer. On the other hand, the second power supply system is the power supply system for sensors (the sensor is the main body, and may include a control device or the like that operates together with the sensor).

[0069] The first power supply system controls the mode by controlling the frequency of the operating clock or the like. On the other hand, the second power supply system controls the mode by means of a relay or a transistor or the like. By dividing the power supply system in this way, the mode can be finely controlled for each power supply system.

[0070] In the above example, the initial operation is detected immediately after the start of operation of the pump device 1 (that is, the trigger signal SG2 is output by capturing the start of operation of the pump device 1). However, the trigger signal SG2 may be output before the start of operation of the pump device 1. For example, when starting operation, the pump device 1 outputs a signal or the like notifying the start of operation. The trigger signal SG2 may be output based on this notification. In this case, the trigger signal SG2 may be output at a timing earlier than the start of operation of the pump device 1.

[0071] Note that in a system configuration where the pump device 1 or the like is managed by a centralized control system, the centralized control device may output the trigger signal SG2.

[0072] As described above, it is desirable that the hardware that realizes the functions of the first detection unit and the communication unit be the power consumption part, and that the first acceleration sensor 2H1 and the communication module 2H3 be switched to the sleep state. Note that the power consumption part may include a part of the control circuit board 2H2 (mainly the part related to the first acceleration sensor 2H1). That is, when the control circuit board 2H2 enters the sleep state, it stops some functions to reduce the power consumption, and it is desirable that at least the functions that can perform mode control to start from the sleep state to the active state when receiving a trigger are maintained.

[0073] Also, in a configuration with the output device 4, it is desirable to include the output device 4 in the power consumption part. That is, it is desirable that the output device 4 be switched to the sleep state together with other power consumption parts.

[0074] When the first acceleration sensor 2H1, the control circuit board 2H2, the communication module 2H3, and the output device 4 are in the active state, that is, when performing the first detection, communication, and output, the power consumption tends to increase. Therefore, if the hardware that realizes the functions of the first detection unit, the determination unit, and the communication unit can be put into the sleep state, the power consumption can be further reduced.

[0075] On the other hand, in the active state, power is "ON" for all of the first acceleration sensor 2H1, the control circuit board 2H2, the communication module 2H3, and the output device 4. Specifically, the active state is a state in which devices such as the first acceleration sensor 2H1 can execute all functions, or a state in which higher functionality is exhibited compared to the sleep state. For example, when the operating clock is set to a low frequency when transitioning to the sleep state, the operating clock is returned to the original high frequency, and the switch from the sleep state to the active state is performed.

[0076] That is, when the sleep state is released, each hardware component that constitutes the power consumption part enters the active state.

[0077] Therefore, when in the sleep state, the power consumption is always lower compared to the case of being in the startup state, and power saving can be achieved.

[0078] The switching between modes such as the sleep state and the startup state is performed via the network. Specifically, when a so-called "magic packet" (that is, a token or the like for switching the mode) is transmitted from an external device to the sensor unit 2 via the network, the sensor unit 2 switches from the sleep state to the startup state. This is the so-called "wake-up".

[0079] In addition, a microcomputer or the like may be configured to receive a "High" or "Low" signal at an input terminal that can be received even in the sleep state.

[0080] Also, in the case of receiving from a digital signal, etc., it may be the case that an analog signal such as a magnetic field is converted into a digital signal.

[0081] Furthermore, a signal triggered by the energy possessed by the radio wave itself may be generated when the radio wave is received.

[0082] Different from the continuously operating pump device 1, a device that operates irregularly has a long time in the stopped state. Different from a device that operates constantly or periodically, a device that operates irregularly is a device whose operation timing is not scheduled in advance.

[0083] On the other hand, in the case of the pump device 1, it is often difficult to obtain data for finding abnormalities and signs of abnormalities even when measuring in the stopped state. That is, when the information processing system 10 can measure the operating pump device 1, it can accurately determine abnormalities and signs of abnormalities.

[0084] Therefore, when the pump device 1 is in a stopped state, it is desirable for the information processing system 10 to be in a sleep state. On the other hand, when the pump device 1 is in an operating state, it is desirable for the information processing system 10 to be in an active state. However, since the schedule is not determined like a device that operates constantly or periodically, it is difficult to set the timing for switching the mode using a timer or the like. Therefore, the information processing system 10 performs mode control by detecting the initial operation of the pump device 1.

[0085] In this way, if the sleep state and the active state can be switched according to the operating time of the pump device 1, the power consumption in the information processing device for IoT can be reduced.

[0086] [Examples of determination of abnormalities and signs of abnormalities] When the pump device 1 operates, an actuator such as a motor operates. And the actuator generates various vibrations or accelerations particularly during operation.

[0087] For example, abnormalities and signs of abnormalities are determined based on Artificial Intelligence (artificial intelligence, hereinafter referred to as "AI"), pre-set rules, or a combination of these. Hereinafter, data indicating the determination results of abnormalities and signs of abnormalities is referred to as "determination data".

[0088] FIG. 5 is a network diagram showing an example of AI. When determination is made by AI, for example, AI having a configuration shown by the following network is used. Hereinafter, the learning model and the learned model are assumed to be implemented on the server 3, that is, on the cloud. However, part or all of the AI may be implemented in an information processing device installed around the sensor unit 2, a so-called "edge server", or the like.

[0089] The configuration of AI (hereinafter referred to as "network 300") has, for example, an input layer L1, an intermediate layer L2 (also referred to as a "hidden layer", etc.), and an output layer L3, etc. That is, the network 300 is configured to perform deep learning or the like.

[0090] The input layer L1 is a layer for inputting data.

[0091] The intermediate layer L2 converts the data input in the input layer L1 based on weights (for example, coefficients used for multiplication) and biases (for example, adding constants). The result processed by the intermediate layer L2 in this way is transmitted to the output layer L3.

[0092] The output layer L3 is a layer for outputting output contents and the like.

[0093] Then, through learning, the weight coefficients (for example, the numerical values input to the data, the characters to be input, or the coefficients for the images are changed based on learning), and parameters to be changed in learning are optimized. Note that the network 300 is not limited to the illustrated network structure. That is, the AI may be realized by other machine learning.

[0094] For example, the AI may be configured to perform preprocessing such as dimensionality reduction (for example, a process of changing a relationship of three dimensions or more to a relationship obtainable by a simple calculation of about three dimensions or less) by machine learning without a teacher. The relationship between the input and the output is preferably processed by a simple calculation such as a linear equation. With such a calculation, the calculation cost can be reduced and the degradation degree can be determined with high accuracy.

[0095] Also, the AI may perform a process for reducing overfitting (also referred to as "overfitting" or "hyperfitting") such as dropout. Other preprocessing such as dimensionality reduction and normalization may also be performed.

[0096] The AI may have a network structure such as a CNN (Convolution Neural Network). Additionally, for example, the network structure may have a configuration such as an LLM (Large Language Model), an RNN (Recurrent Neural Network), or an LSTM (Long Short-Term Memory). That is, the AI may have a network structure other than deep learning.

[0097] Also, the AI may have a configuration with hyperparameters. That is, the AI may be configured such that some settings are performed by the user or the like. Furthermore, the AI may identify the features to be learned, or the user may set some or all of the features to be learned.

[0098] And the AI may utilize other machine learning. For example, the AI may perform preprocessing such as normalization using an unsupervised model. Furthermore, the learning may be, for example, reinforcement learning (a learning method in which the AI is made to make a selection and an evaluation (reward) for the selection is given to increase the evaluation).

[0099] In learning, data augmentation or the like may be performed. That is, preprocessing may be performed to expand one experimental data or the like into a plurality of learning data in order to increase the learning data used for learning the learning model. In this way, if the learning data can be increased, the learning of the learning model can be further advanced.

[0100] In addition, the AI may be configured to perform transfer learning, fine-tuning, or the like. That is, since the execution environment often varies from device to device for information processing devices using AI, the settings may differ for each device according to the execution environment. For example, the basic configuration of the AI is learned by another information processing device. Thereafter, each information processing device may be additionally learned or configured, etc., in order to further optimize for each execution environment.

[0101] When making a determination based on rules, the determination is realized by presetting a relational expression indicating the relationship between the input (i.e., sensor data) and the output (i.e., the determination result), or a table (the so-called "look-up table (LUT)").

[0102] Note that the AI and the rules may be combined. For example, the AI and the rules may be combined in terms of the type of the pump device 1, the content of the sensor data, or redundant determination to generate determination data.

[0103] The determination is not limited to whether or not an abnormality, i.e., a state where a failure or the like has already occurred in the pump device 1, but also determines a sign of an abnormality, which is a state before the occurrence of the abnormality.

[0104] Note that the criteria for determining an abnormality are preferably not uniform but individually set according to the type of the pump device 1, etc. For example, even when there are multiple pump devices 1 of the same type (for example, when the manufacturer, model, etc. are the same), it is desirable to set different determination criteria for each pump device 1.

[0105] Hereinafter, an example will be given in which the first device and the second device of the same type of the pump device 1 are installed and operating in a factory. Even for the first device and the second device of the same type, for example, if the installation date or the start date of operation is different, it is often possible to more accurately determine an abnormality by setting different determination criteria.

[0106] For example, assume that the first device tends to generate a lot of vibrations even when it is normal. If such a first device is judged according to the same judgment criteria as the second device, the first device is likely to be misdetected as abnormal even when it is normal. Therefore, it is desirable that the judgment criteria for the first device be a so-called "loose" judgment criteria that judges it as normal even if there are many vibrations compared to the second device.

[0107] On the other hand, for the second device that normally has few vibrations, it is desirable to have a so-called "strict" judgment criteria that judges it as abnormal even for a little vibration.

[0108] Thus, even for pump devices 1 of the same type, individual devices often have idiosyncrasies. Therefore, it is desirable that the judgment criteria be not uniform but set separately for each pump device 1. With such a setting of judgment criteria, it is possible to cope with the idiosyncrasies of individual devices and detect abnormalities accurately.

[0109] Also, it is desirable that the judgment criteria can be updated. For example, in the case of pump device 1 that tends to generate a lot of vibrations even when it is normal (hereinafter referred to as "before repair"), such idiosyncrasies may disappear after repair or the like. Therefore, it is desirable that the judgment criteria can be updated after an event such as repair. Note that the event is not limited to repair and may be a setting change of pump device 1 or the like. Hereinafter, an example where the event is repair will be described.

[0110] Specifically, the judgment criteria for pump device 1 (hereinafter referred to as "before repair") that tends to generate a lot of vibrations corresponds to the state before repair. Therefore, if pump device 1 after repair (hereinafter referred to as "after repair") that has been repaired so that the vibrations are reduced is judged according to the judgment criteria for before repair, it is easy to misdetect an abnormality.

[0111] Therefore, it is desirable that different judgment criteria can be set before and after repair. For pump device 1, the idiosyncrasies and the like may change significantly before and after repair. Therefore, if the same judgment criteria are used before and after the event, it will be judged according to the judgment criteria considering the same idiosyncrasies. Therefore, it is desirable that the judgment criteria can be updated in response to events and the like.

[0112] In this way, if the determination criteria can be updated, even if the characteristics of the pump device 1 change significantly due to an event such as repair, abnormalities can be detected accurately.

[0113] A failure means a state in which an item has lost its ability to perform as required, as defined in JIS Z8115:2019. Also, the time during which a failure occurs becomes the downtime. Furthermore, a failure includes both a wear failure (also referred to as a "deterioration failure") and an aging failure (also referred to as an "age-related failure") in JIS Z8115:2019.

[0114] An omen of an abnormality corresponds to an abnormal state, that is, a state before the pump device 1 becomes inoperable. Therefore, when there is an omen of an abnormality, the pump device 1 is still in an operable state but is in a state where an abnormality is likely to occur soon.

[0115] An omen of an abnormality is determined by vibration. And the omen of an abnormality may be determined by the vibration in the audible range generated during the operation of the pump device 1, that is, "sound". In a state before an abnormality occurs, the pump device 1 often generates a sound that is not emitted in a normal state, that is, a so-called "abnormal sound". For example, a maintenance worker with a lot of experience can accurately determine an omen of an abnormality by distinguishing the abnormal sound. Therefore, if the vibration in the audible range is used for determination, the information processing system 10 can accurately determine an omen of an abnormality separately from an abnormality.

[0116] On the other hand, the vibration may be detected with respect to the floor on which the pump device 1 is installed or the floor connected to the installation position of the pump device 1.

[0117] The vibrations in the audible range have a frequency band of 20 Hz to 20 kHz. Therefore, to sense the vibrations in the audible range, a configuration using a microphone as a sensor may be adopted. That is, by using a microphone as a sensor, sensor data mainly collecting voices may be generated. Then, based on whether the sensor data contains abnormal sounds, AI or the like can accurately determine abnormalities and signs of abnormalities.

[0118] When signs of abnormalities can be determined, instead of "corrective maintenance (JIS Z8115:2019)" which responds after an abnormality occurs, "condition-based maintenance" of "preventive maintenance (JIS Z8115:2019)" which responds before an abnormality occurs can be accurately performed. Or, when signs of abnormalities can be determined, preparations such as preparing parts in advance can be made so that an abnormality can be detected in advance and "corrective maintenance" can be promptly performed when a failure occurs. In this way, if replacement parts or substitutes for the main body can be prepared in advance, the impact on the factory operation caused by the pump device 1 stopping can be prevented.

[0119] As described above, the information processing system 10 determines abnormalities and signs of abnormalities. In addition, the information processing system 10 determines a state that is neither an abnormality nor a sign of an abnormality as "normal". Therefore, the information processing system 10 outputs determination results including abnormalities, signs of abnormalities, and normal in at least three levels in a hierarchical manner.

[0120] Note that depending on the type of the pump device 1, there may be a plurality of abnormalities and signs of abnormalities. In such a case, it is desirable that the abnormalities and signs of abnormalities be determined by being classified into multiple types.

[0121] Abnormalities and signs of abnormalities may be further subdivided into multiple levels. For example, abnormalities may have multiple levels, such as "severe abnormalities" with a high degree of severity like a failure, and "minor abnormalities" with a low degree of severity of the failure. Therefore, the hierarchical output may be four levels or more.

[0122] In this way, if the state can be determined not only for abnormalities and normal conditions but also for signs of abnormalities, the pump device 1 can be made highly reliable by preventive maintenance and the like.

[0123] Also, if a maintenance worker, that is, a human being, tries to determine abnormalities and signs of abnormalities by vibration, high experience is required. On the other hand, since the information processing system 10 determines based on AI or rules, it can accurately determine abnormalities and signs of abnormalities without relying on a maintenance worker with high experience.

[0124] [Overall Processing Example] FIG. 6 is a diagram showing an overall processing example.

[0125] In step S01, the information processing system 10 performs mode control to put power-consuming parts such as the output device 4 into a sleep state. That is, when the generation of sensor data and the like are completed, the information processing system 10 makes a mode transition to the sleep state.

[0126] Note that depending on the pump device 1 or the information processing system 10, there may be an initialization process. For example, when the information processing system 10 is installed for the first time, an initialization process of temporarily operating the pump device 1 to perform various checks may be first executed. And in the initialization process, the information processing system 10 performs, for example, various checks, settings, or data acquisition. After such an initialization process, the information processing system 10 may perform mode control to put power-consuming parts such as the output device 4 into a sleep state. Also, the initialization process may be executed at an arbitrary timing based on, for example, a user's operation or the like, in addition to the case when the information processing system 10 is installed for the first time.

[0127] In step S02, the information processing system 10 maintains the mode in the sleep state.

[0128] In step S03, the information processing system 10 determines whether the pump device 1 has started operating. That is, when the operation of the pump device 1 is detected, the information processing system 10 outputs a trigger (YES in step S03). On the other hand, when the operation of the pump device 1 is not detected, the information processing system 10 does not output a trigger and continues to detect the start of operation of the pump device 1 (NO in step S03).

[0129] In step S04, the information processing system 10 performs mode control to activate power-consuming parts such as the output device 4. Thereafter, in the overall process, the information processing system 10 is in an activated state.

[0130] In step S05, the information processing system 10 measures the pump device 1 by the sensor unit 2. Accordingly, sensor data indicating the measurement result is generated.

[0131] In step S06, the information processing system 10 transmits the sensor data from the sensor unit 2 to the server 3 by wireless communication. And the server 3 receives the sensor data.

[0132] In step S07, the information processing system 10 determines an abnormality in the pump device 1 and a sign of the occurrence of the abnormality based on the sensor data. Accordingly, determination data indicating the determination result is generated.

[0133] In step S08, the information processing system 10 transmits the determination data from the server 3 to the output device 4. And the output device 4 receives the determination data. In the case where the sensor unit 2 controls the output device 4, etc., the transmission destination may be the sensor unit 2. That is, the transmission destination of the determination data may be any device that controls the output device 4 to output based on the determination data.

[0134] In step S09, the information processing system 10 causes the output device 4 to output the determination result step by step.

[0135] [Functional configuration example] FIG. 7 is a diagram showing a functional configuration example. For example, the information processing system 10 includes a first detection unit 10F1, a second detection unit 10F2, a determination unit 10F3, a communication unit 10F4, and a control unit 10F5. Further, it is desirable that the information processing system 10 further includes an output unit 10F6.

[0136] The first detection unit 10F1 performs a first detection procedure for detecting the vibration or acceleration of the pump device 1. For example, the first detection unit 10F1 is realized by a first acceleration sensor 2H1 or the like.

[0137] The second detection unit 10F2 performs a second detection procedure for detecting whether the pump device 1 is operating or stopped. For example, the second detection unit 10F2 is realized by a second acceleration sensor 2H5 or the like.

[0138] The determination unit 10F3 performs a determination procedure for determining a state including a sign of abnormality of the pump device 1 based on the detection data and generating determination data. For example, the determination unit 10F3 is realized by a CPU 3H1 or the like.

[0139] The communication unit 10F4 performs a communication procedure for transmitting or receiving the detection data or the determination data. For example, the communication unit 10F4 is realized by a communication module 2H3 or the like.

[0140] The control unit 10F5 has two or more control modes and performs a control procedure for switching the control mode according to the detection result by the second detection unit 10F2 and controlling the first detection unit 10F1 and the communication unit 10F4. For example, the control unit 10F5 is realized by a control circuit board 2H2 or the like.

[0141] The output unit 10F6 performs an output procedure for outputting the state in three or more levels based on the determination data. For example, the output unit 10F6 is realized by an output device 4 or the like.

[0142] In the case where the pump device 1 that operates irregularly is the target for determining the state, when the pump device 1 is not operating, the information processing system 10 mainly puts the first detection unit 10F1 and the communication unit 10F4 into the sleep state. On the other hand, in order to detect the operation of the pump device, when the information processing system 10 detects the operation of the pump device 1, it can mainly activate the first detection unit 10F1 and the communication unit 10F4 according to the operation of the pump device 1.

[0143] When the first detection unit 10F1 and the communication unit 10F4 are in the activated state, the power consumption tends to increase. Therefore, when the information processing system 10 can put the first detection unit 10F1 and the communication unit 10F4, which consume a large amount of power, into the sleep state, it can reduce the power consumption in the information processing device for IoT.

[0144] Similarly, in the configuration including the output unit 10F6, since the output device 4 also often consumes a large amount of power, it is desirable that it can be put into the sleep state together with the first detection unit 10F1 and the communication unit 10F4.

[0145] On the other hand, when the pump device 1 that operates irregularly is the target, it cannot be handled by mode control such as starting in advance according to the operation time with a timer. Therefore, if the second detection unit 10F2 can detect the operation of the pump device, even when the pump device 1 that operates irregularly is the target, the power consumption can be suppressed and the state can be determined according to the operation.

[0146] Especially when using IoT in factories and the like, it is difficult to supply power by wire (such as a configuration that supplies power from a so-called power outlet), and in many cases, the power is supplied by a battery 2H4 or the like. Therefore, if devices with a large power consumption such as sensors and output devices are in the activated state for a long time, the overall power supply time tends to be short. On the other hand, if the control can switch between the activated state and the sleep state according to the operation time of the pump device 1 as in the above configuration, the information processing system 10 can significantly reduce the power consumption in IoT.

[0147] Also, if the power consumption can be reduced, the number of replacements of the battery 2H4 and the like can be reduced. Therefore, the information processing system 10 can improve maintainability.

[0148] As described above, in the configuration of transmitting and receiving determination data by communication, compared with the case of performing it by wire, it is possible to reduce the risk of the user being involved in the rotating device and improve safety. Further, the output device 4 outputs the determination result so that the user can know it even from a position keeping a distance, so that the risk of the user being involved in the rotating device or the like is reduced by notification by light, sound, or communication, thereby further improving safety.

[0149] Therefore, if the information processing system 10 can reduce the power consumption, it can extend the operation time and improve the reliability.

[0150] [User approach detection and output examples] Sensing by the sensor unit 2 and generation of sensor data are performed in accordance with the operation time of the pump device 1. On the other hand, it is desirable to output the determination result in accordance with the approach of the user 5.

[0151] Approach means that the user 5 is within a certain distance from the output device 4.

[0152] FIG. 8 is a diagram showing an example of output based on the approach of the user. For example, the approach of the user 5 is detected by a human sensor 6 or the like.

[0153] The human presence sensor 6 is a communication-based detector, an optical sensor, an infrared sensor, a pressure sensor, a vibration sensor, a position sensor, or the like. It is desirable that the human presence sensor 6 also serves as the sensor included in the sensor unit 2. That is, the sensor unit 2 preferably has a configuration for detecting the approach of the user 5. For example, if the frequency sensed by the sensor unit 2 includes the audible range, the sensor unit 2 detects the approach of the user 5 by detecting footsteps due to the approach of the user 5, vibrations due to walking, or voices. In this way, if the sensor unit 2 also serves as the human presence sensor 6, the number of sensors can be reduced. However, the human presence sensor 6 does not have to be integrated with the sensor unit 2. Note that the human presence sensor 6 may be an external device or have a separate power supply. That is, the human presence sensor 6 may have a power supply configuration in which power is supplied from a factory outlet or a switchboard instead of the battery 2H4.

[0154] Also, when the human presence sensor 6 is configured as an external device, when the human presence sensor 6 senses the approach of a person, the human presence sensor 6 transmits a detection signal notifying the detection to the information processing system 10. In the case of such a detection signal from the human presence sensor 6, the information processing system 10 performs the operation without omitting the output by the output device 4.

[0155] Also, it is desirable that the approach of the user 5 be detected when the user 5 approaches the output device 4. For example, when the vibration generated by the user 5 increases with time or the distance to the user 5 with respect to the output device 4 decreases, the human presence sensor 6 detects that the user 5 is approaching the output device 4.

[0156] When the user 5 approaches the output device 4, it is highly likely that the user 5 is trying to check the state of the pump device 1. Therefore, by aiming for the case where the user 5 approaches, if the output device 4 performs an output to the user 5, it is possible to prevent the user 5 from missing it.

[0157] Incidentally, it is desirable that the output device 4 includes a storage device that stores determination data. As described above, the output device 4 may receive the determination data and not immediately output the determination result to notify the user 5. In such a case, it is desirable that the output device 4 stores the determination data and can hold the determination result until the approach of the user 5.

[0158] In this way, in order for sensing, determination, and output to be executed at optimal timings, it is desirable that each device starts up. When each device starts up at the timings as described above, the time during which each device is in the startup state is shortened, and the information processing system 10 can reduce the power consumption more.

[0159] [Modification Example of the Second Detection Unit] The second detection unit 10F2 may be configured to use a type other than the acceleration sensor. That is, the second detection unit 10F2 may detect the operation of the pump device 1 other than acceleration or vibration.

[0160] The second detection unit 10F2 may measure the voltage, current, sound, electric field, magnetic field, or a combination thereof generated in the pump device 1 to detect the operation of the pump device.

[0161] For example, when the pump device 1 starts up, it often generates a large voltage, current, sound, electric field, or magnetic field. Specifically, compared with the steady operation, the startup causes rotation by a force greater than the static friction force, operates at a higher rotation speed than the steady operation, or passes a larger current than the steady operation, such as by charging a capacitor, or performs startup-specific operations not performed in the steady operation, or often uses specific startup values. Also, at startup, the voltage may increase or a large magnetic field may be generated to supply a large current. Therefore, it is desirable that the second detection unit 10F2 is configured to detect these specific phenomena that occur during startup. Therefore, the type of sensor that realizes the second detection unit 10F2 is a type suitable for the phenomena that occur during startup.

[0162] For example, when detecting the initial movement by sound, the second detection unit 10F2 is realized as a sensor such as a microphone or an acceleration sensor having a sensitivity band in the audible range.

[0163] Similarly, when detecting by voltage, current, electric field, magnetic field, etc., the second detection unit 10F2 is realized as a sensor such as a voltmeter, ammeter, electric field sensor, or magnetic field sensor.

[0164] In addition, since it is only necessary for the second detection unit 10F2 to be able to detect the operation of the pump device 1, it may be configured to send a signal when the pump device 1 operates. Therefore, the second detection unit 10F2 may be configured to output a trigger in response to the initial movement signal sent by the pump device 1.

[0165] Note that the second detection unit 10F2 may be composed of multiple types or multiple sensors. For example, the second detection unit 10F2 may detect the initial movement with sensors having a redundant configuration. In this way, if the sensors are redundant, that is, the trigger is output based on the result of a so-called "AND" calculation of the initial movement detection results, the initial movement detection results can be accurately detected. On the other hand, in order to reduce missing the timing of acquiring data, a configuration that performs a so-called "OR" calculation may also be used.

[0166] [Other Embodiments] The above processing and the data used in the processing executed in this embodiment may be executed and stored by an information processing system. For example, the information processing system may execute or store with multiple information processing devices in order to realize redundancy, dispersion, parallelism, or a combination thereof. Therefore, the present invention may be realized by a device other than the hardware configuration shown above and a system other than the device shown above.

[0167] The above control method may also be realized by a program that causes a computer to execute overall processing and the like.

[0168] Furthermore, the program according to the present invention is not limited to a single program and may be an aggregate of multiple programs. Also, the program according to the present invention is not limited to being executed on a single device and may be executed in a distributed manner by multiple information processing devices. Furthermore, the role sharing among the respective information processing devices is not limited to the above-described example. That is, part or all of the above-described processing may be executed by an information processing device different from the above-described information processing devices.

[0169] Part or all of each means realized by the program can also be realized by hardware such as an integrated circuit. Furthermore, the program may be provided by being recorded on a non-transitory computer-readable recording medium. The recording medium refers to, for example, a hard disk, an SD card (registered trademark), an optical disk such as a DVD, or a server on the Internet. Therefore, the program may be distributed via an electric communication line such as the Internet.

[0170] Also, the information processing devices and the like constituting the information processing system may be located overseas.

[0171] Note that the present invention is not limited to the respective embodiments exemplified above. Therefore, the present invention can be added to or modified in its components without departing from the technical gist. Thus, all of the technical matters included in the technical idea described in the claims are the subject of the present invention. Note that the embodiments exemplified above are suitable specific examples in practice. And those skilled in the art can realize various modification examples from the disclosed content, and such modification examples are included in the technical scope described in the claims.

[0172] [Contribution to the Promotion of SDGs] As described above, the present invention has the effect of saving power. As a result, it contributes to "Goal 7: Ensure access to affordable, reliable, sustainable and modern energy for all" in the SDGs.

Explanation of Reference Numerals

[0173] 1: Pump device 2: Sensor Unit 2H1: First Acceleration Sensor 2H2: Control Circuit Board 2H3: Communication Module 2H4: Battery 2H5: Second Acceleration Sensor 3: Server 4: Output Device 5: User 6: Presence Sensor 10: Information Processing System 10F1: First Detection Unit 10F2: Second Detection Unit 10F3: Judgment Unit 10F4: Communication Unit 10F5: Control Unit 10F6: Output Unit SG1: Status Signal SG2: Trigger Signal SG3: Mode Signal T1: First Timing T2: Second Timing T3: Third Timing T4: Fourth Timing

Claims

1. A first detection unit that detects vibration or acceleration of a rotating device that operates irregularly; A second detection unit that detects whether the rotating device is operating or stopped; A determination unit that determines a state including an abnormal sign of the rotating device based on detection data indicating a detection result by the first detection unit and generates determination data; A communication unit that transmits or receives the detection data or the determination data; An information processing apparatus comprising a control unit that has two or more control modes, switches the control mode according to the second detection unit, and controls the first detection unit and the communication unit .

2. The second detection unit detects voltage, current, sound, electric field, magnetic field, or a combination thereof generated in the rotating device The information processing apparatus according to claim 1.

3. The control mode includes a first control mode that controls the first detection unit or the communication unit to an activated state, and a second control mode that controls the first detection unit or the communication unit to a power-saving state The information processing apparatus according to claim 1.

4. The control unit switches from the second control mode to the first control mode based on a trigger output by the second detection unit The information processing apparatus according to claim 3.

5. The second detection unit has a first detection mode for detecting the operation or stop of the rotating device, and a second detection mode for not detecting the operation or stop of the rotating device, and The control unit switches between the first detection mode and the second detection mode The information processing apparatus according to claim 1.

6. The control unit switches between the first detection mode and the second detection mode at a predetermined period The information processing apparatus according to claim 5.

7. The second detection mode is a state in which power supply to the second detection unit is stopped The information processing apparatus according to claim 6.

8. The determination unit uses a learned model learned with learning data including the detection data and the determination data The information processing apparatus according to claim 1.

9. The determination unit makes a determination using a table showing the relationship between the detection data and the determination data The information processing apparatus according to claim 1.

10. The information processing apparatus further includes an output unit that outputs the state in three or more levels based on the determination data, and The control unit switches from the second control mode to the first control mode The information processing apparatus according to claim 3.

11. detects the approach of a user, and The output unit When the approach of the user is detected, output the state step by step. The information processing apparatus according to claim 10.

12. Store the determination data. The output unit Based on the most recent determination data, output the state step by step. The information processing apparatus according to claim 10.

13. A first detection unit that detects the vibration or acceleration of a rotating device that operates irregularly, A second detection unit that detects whether the rotating device is operating or stopped, A determination unit that determines a state including an omen of abnormality of the rotating device based on detection data indicating the detection result by the first detection unit and generates determination data, A communication unit that transmits or receives the detection data or the determination data, A control unit that has two or more control modes, switches the control mode according to the second detection unit, and controls the first detection unit and the communication unit An information processing system comprising.

14. A first detection unit that detects the vibration or acceleration of a rotating device that operates irregularly, An information processing apparatus comprising a communication unit that transmits and receives determination data A program for causing the information processing apparatus to execute a control method, A second detection procedure for detecting whether the rotating device is operating or stopped, A determination procedure for determining a state including an omen of abnormality of the rotating device based on detection data indicating the detection result by the first detection unit and generating the determination data, A control procedure that has two or more control modes, switches the control mode according to the result of the second detection procedure, and controls the first detection unit and the communication unit A program for causing the above to be executed.

Images