Interface device, interface method, and recording medium having recorded interface program

Abstract

The present invention provides an interface device, an interface method and a recording medium having an interface program recorded thereon, wherein the interface device comprises: an interface unit, which receives a first instruction using a general format protocol that is independent of a cloud platform that manages the device from an AI processing unit that performs at least one of a process of generating a model for determining the state of a device through machine learning or a process of using the model to determine the state of a device; and a cloud communication unit, which converts the first instruction into a second instruction using a first format protocol that is dependent on a first cloud platform, and sends the second instruction to the first cloud platform.

Description

Technical Field

[0001] The invention relates to an interface device, an interface method and a recording medium recording an interface program. Background Art

[0002] Patent Document 1 states that “a diagnostic device 100 generates an analysis model for diagnosing abnormal operation of a device 10 based on detection signals from a plurality of sensors 20 and improves the accuracy of diagnosis by learning the analysis model.”

[0003] Prior art literature

[0004] Patent Document 1: Japanese Patent Publication No. 2019-101495 Summary of the invention

[0005] In a first embodiment of the present invention, an interface device is provided. The interface device may include an interface unit that receives a first instruction using a general format protocol that is independent of a cloud platform that manages the device from an AI processing unit that performs at least one of a process of generating a model for determining the state of a device through machine learning or a process of determining the state of a device using a model. The interface device may include a cloud communication unit that converts the first instruction into a second instruction using a first format protocol that is dependent on a first cloud platform, and sends the second instruction to the first cloud platform.

[0006] The cloud communication unit may receive a third instruction using the first format protocol from the first cloud platform, and convert the third instruction into a fourth instruction using the universal format protocol, and the interface unit sends the fourth instruction to the AI ​​processing unit.

[0007] The cloud communication unit may include a plurality of inherent communication units corresponding to the plurality of cloud platforms, respectively.

[0008] The first instruction may include a sensor list acquisition request instruction for requesting to acquire a list of sensors set in the device managed by the first cloud platform.

[0009] The first instruction may include a sensor data acquisition request instruction for requesting to acquire measurement data of a sensor provided in a device managed by the first cloud platform.

[0010] The first instruction may include a judgment result write request instruction for requesting to write the judgment result of the status of the device into the database of the first cloud platform.

[0011] The first instruction may include a model information change request instruction for requesting to change information of a model managed by the first cloud platform.

[0012] The first instruction may include a configuration information change request instruction for requesting to change configuration information managed by the first cloud platform.

[0013] The interface device may further include an AI processing unit.

[0014] The AI ​​processing unit may include a learning unit that generates a model through machine learning.

[0015] The AI ​​processing unit may include a determination unit that determines the state of the device using the model.

[0016] In a second embodiment of the present invention, an interface method is provided. The interface method may include receiving a first instruction using a general format protocol that is independent of a cloud platform that manages the device from an AI processing unit that performs at least one of a process of generating a model for determining the state of a device through machine learning or a process of using a model to determine the state of a device. The interface method may include converting the first instruction into a second instruction using a first format protocol that is dependent on a first cloud platform, and sending the second instruction to the first cloud platform.

[0017] In a third embodiment of the present invention, a recording medium having an interface program recorded thereon is provided. The interface program can be executed by a computer. The computer can function as an interface unit by executing the interface program, and the interface unit receives a first instruction using a general format protocol that is independent of a cloud platform that manages the device from an AI processing unit that performs at least one of a process of generating a model for determining the state of a device through machine learning or a process of determining the state of a device using a model. The computer can function as a cloud communication unit by executing the interface program, and the cloud communication unit converts the first instruction into a second instruction using a first format protocol that is dependent on the first cloud platform, and sends the second instruction to the first cloud platform.

[0018] In addition, the above summary of the invention does not list all the essential features of the present invention. In addition, sub-combinations of these feature groups may also constitute inventions. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 An example of a block diagram of the interface device 100 according to the present embodiment is shown together with the device 10 and the cloud platform 50 .

[0020] Figure 2 An example of a flow of generating a learning model using the interface device 100 according to the present embodiment is shown.

[0021] Figure 3 An example of a determination flow of a learning model using the interface device 100 according to the present embodiment is shown.

[0022] Figure 4 An example of a flow of updating a learning model from the external system 300 using the interface device 100 according to the present embodiment is shown.

[0023] Figure 5Another example of the flow of updating the learning model from the external system 300 using the interface device 100 of the present embodiment is shown.

[0024] Figure 6 The computer 9900 is an example that can implement various aspects of the present invention in whole or in part.

[0025] Description of Reference Numerals

[0026] 10 equipment, 20 sensors, 50 cloud platform, 100 interface device, 110 interface unit, 120 cloud communication unit, 200 AI processing unit, 210 learning unit, 220 learning model generation unit, 230 learning model management unit, 240 judgment unit, 250 judgment processing unit, 9900 computer, 9901 DVD-ROM, 9910 main controller, 9912 CPU, 9914 RAM, 9916 graphics controller, 9918 display device, 9920 input / output controller, 9922 communication interface, 9924 hard disk drive, 9926 DVD-ROM drive, 9930 ROM, 9940 input / output chip, 9942 keyboard. DETAILED DESCRIPTION

[0027] The present invention will be described below by way of the embodiments of the invention, but the following embodiments do not limit the invention according to the claims. In addition, not all combinations of features described in the embodiments are essential to the solution means of the invention.

[0028] Figure 1 An example of a block diagram of the interface device 100 of the present embodiment is shown together with the device 10 and the cloud platform 50. The interface device 100 of the present embodiment provides an interface between a functional unit that performs AI (Artificial Intelligence) processing related to the state judgment (e.g., abnormality diagnosis) of the device 10 and the cloud platform 50 that manages the device 10. At this time, the interface device 100 receives instructions using a general format protocol that is not dependent on the cloud platform 50 from the functional unit that performs the AI ​​processing. In addition, the interface device 100 converts the instructions using the general format protocol into instructions using an inherent format protocol that depends on an arbitrary cloud platform, and sends the converted instructions to an arbitrary cloud platform.

[0029] The device 10 is a facility that is the object of the state judgment of the functional unit that performs AI processing. The device 10 can be, for example, a factory. In addition to industrial factories such as chemical factories, such factories can also be factories that manage and control wellheads such as gas fields or oil fields and their surroundings, factories that manage and control power generation such as hydropower, thermal power, and atomic energy, factories that manage and control environmental power generation such as solar power or wind power, or factories that manage and control water supply and drainage or dams, etc. However, it is not limited to this. The device 10 can also be any facility different from a factory. The device 10 is provided with one or more sensors 20.

[0030] The sensor 20 measures the physical quantity of the object. For example, the sensor 20 may be an IIoT (Industrial Internet of Things) sensor that is later installed in the factory. As an example, the sensor 20 may be able to obtain measurement data of physical quantities such as vibration, temperature, humidity, illumination, odor, gas concentration, pressure, air pressure, stress, magnetic field, sound or image in the factory. In addition, in the above description, as an example, the case where the sensor 20 is an IIoT sensor that is later installed in the factory is shown, but it is not limited to this. For example, the sensor 20 may be a process control (measurement) sensor pre-installed in the OT (Operational Technology) area, or an industrial sensor that is connected to or integrally formed with one or more field devices installed in the factory. The measurement data measured by the sensor 20 is collected to the cloud platform 50, for example, via a wireless module (not shown) installed in the device 10, and is managed on the cloud platform 50.

[0031] The cloud platform 50 is a cloud system provided in the device 10. The cloud platform 50 has a function unit for executing AI processing and a function required for constructing a state determination system of the device 10. For example, the cloud platform 50 may have a function for storing measurement data of the sensor 20 collected via a wireless module in a database. In addition, the cloud platform 50 may have a function for starting and executing applications in the cloud.

[0032] Here, there may be multiple cloud systems with different specifications for each operator providing cloud services, but the type of cloud set up in the device 10 can be arbitrary. In this figure, as an example, a cloud platform 50 is shown with a first cloud platform 50A provided by operator A and a second cloud platform 50B provided by operator B (collectively referred to as "cloud platform 50"). The first cloud platform 50A, for example, can provide a REST (Representational State Transfer) API (Application Programing Interface) as an interface. In addition, the second cloud platform 50B, for example, can provide GraphQL as an interface. Thus, multiple cloud platforms 50 can provide multiple interfaces with different specifications.

[0033] The interface device 100 of the present embodiment provides an interface between the cloud platform 50 , whose specifications differ for each operator providing cloud services, and a functional unit that executes AI processing.

[0034] The interface device 100 may be a computer such as a PC (personal computer), a tablet computer, a smart phone, a workstation, a server computer, or a general-purpose computer, or a computer system connected to multiple computers. Such a computer system is also a computer in a broad sense. In addition, the interface device 100 may be implemented by one or more virtual computer environments that can be executed in a computer. Instead, the interface device 100 may be a dedicated computer designed for interface use, or may be dedicated hardware implemented by a dedicated circuit. In addition, the interface device 100 may also be implemented by cloud computing. For example, the interface device 100 may be a system attached to the cloud platform 50 by executing an interface program by a computer.

[0035] The interface device 100 includes an interface unit 110 and a cloud communication unit 120. In addition, the interface device 100 may also include an AI processing unit 200. These blocks are functionally separated functional blocks and may not necessarily be consistent with the actual device configuration. That is, although it is represented as one block in this figure, it may not necessarily be composed of one device. In addition, although it is represented as different blocks in this figure, they may not necessarily be composed of different devices.

[0036] In addition, in this figure, as an example, the AI ​​processing unit 200 is shown as a function of a part of the interface device 100 and is integrally configured with the interface unit 110 and the cloud communication unit 120, but the present invention is not limited thereto. A part or all of the AI ​​processing unit 200 may be configured separately from the interface device 100.

[0037] The interface unit 110 receives a first instruction using a universal format protocol that is independent of the cloud platform 50 that manages the device 10 from the AI ​​processing unit 200, and the AI ​​processing unit 200 performs at least one of a process of generating a model for determining the state of the device 10 by machine learning or a process of determining the state of the device 10 using the model. The details of such instructions will be described later. The interface unit 110 supplies the first instruction received from the AI ​​processing unit 200 to the cloud communication unit 120.

[0038] The cloud communication unit 120 communicates with the cloud platform 50. The cloud communication unit 120 may have multiple inherent communication units corresponding to multiple cloud platforms 50. In this figure, as an example, the cloud communication unit 120 has a first cloud communication unit 120A corresponding to the first cloud platform 50A and a second cloud communication unit 120B corresponding to the second cloud platform 50B (collectively referred to as "cloud communication unit 120").

[0039] The first cloud communication unit 120A converts the first instruction for the first cloud platform 50A supplied from the interface unit 110 into an instruction using a protocol in a format corresponding to the REST API provided by the first cloud platform 50A, and sends the converted instruction to the first cloud platform 50A. That is, the cloud communication unit 120 converts the first instruction supplied from the interface unit 110 into a second instruction using a protocol in a first format dependent on the first cloud platform 50A, and sends the second instruction to the first cloud platform 50A. Similarly, the second cloud communication unit 120B converts the first instruction for the second cloud platform 50B supplied from the interface unit 110 into an instruction using a protocol in a format corresponding to the GraphQL provided by the second cloud platform 50B, and sends the converted instruction to the second cloud platform 50B.

[0040] In addition, in the above description, as an example, the interface unit 110 and the cloud communication unit 120 relay instructions from the AI ​​processing unit 200 to the cloud platform 50, and the interface unit 110 and the cloud communication unit 120 can also relay instructions in the opposite direction. That is, the interface unit 110 and the cloud communication unit 120 can also relay instructions from the cloud platform 50 to the AI ​​processing unit 200. For example, the first cloud communication unit 120A can receive a third instruction using a protocol in a format corresponding to the REST API from the first cloud platform 50A. And, the first cloud communication unit 120A can convert the third instruction into a fourth instruction using a universal format protocol and supply it to the interface unit 110. That is, the cloud communication unit 120 can receive a third instruction using a first format protocol from the first cloud platform 50A, and convert the third instruction into a fourth instruction using a universal format protocol. Similarly, the second cloud communication unit 120B can receive a third instruction using a protocol in a format corresponding to GraphQL from the second cloud platform 50B. Furthermore, the second cloud communication unit 120B may convert the third instruction into a fourth instruction using the universal format protocol and supply the fourth instruction to the interface unit 110. Furthermore, the interface unit 110 may send the fourth instruction supplied from the first cloud communication unit 120A or the second cloud communication unit 120B to the AI ​​processing unit 200.

[0041] In addition, in the above description, the cloud communication unit 120 is shown as an example in which the cloud communication unit 120 is interposed between the interface unit 110 and the cloud platform 50 and the cloud communication unit 120 converts the command, but it is not limited to this. In the case where it is not necessary to convert the command for each cloud platform 50, the interface device 100 can also omit the command conversion performed by the cloud communication unit 120. That is, the interface unit 110 and the cloud platform 50 can also communicate directly without going through the cloud communication unit 120.

[0042] The AI ​​processing unit 200 performs at least one of a process of generating a model for determining the state of the device 10 by machine learning or a process of using the model to determine the state of the device 10. Here, as an example, the state determination process of the device 10 may be a device abnormality diagnosis process for diagnosing an abnormality of the device 10. However, this is not limited to this. The state determination process of the device 10 may be expanded to a process involving various industrial solutions, such as a pipe corrosion diagnosis process for diagnosing corrosion of pipes.

[0043] The AI ​​processing unit 200 includes a learning unit 210, a learning model generation unit 220, a learning model management unit 230, a judgment unit 240, and a judgment processing unit 250. In addition, these blocks are functionally separated functional blocks and may not necessarily be consistent with the actual device configuration. That is, although they are represented as one block in this figure, they may not necessarily be composed of one device. In addition, although they are represented as different blocks in this figure, they may not necessarily be composed of different devices.

[0044] The learning unit 210 generates a model for determining the state of the device 10 by machine learning. For example, the learning unit 210 generates a learning model (AI model) for determining the state of the device 10 by performing machine learning using data input from the learning model generation unit 220 .

[0045] The learning model generation unit 220 cooperates with the learning unit 210 to perform new creation or update processing of the model for determining the state of the device 10. For example, the learning model generation unit 220 inputs data obtained from the cloud platform 50 via the interface unit 110 and the cloud communication unit 120 to the learning unit 210, and uses the learning unit 210 to perform new creation or update processing of the learning model.

[0046] The learning model management unit 230 manages information related to the model for determining the state of the device 10. For example, the learning model management unit 230 stores and manages information of the learning model generated by the learning unit 210 and the learning model generation unit 220 together with configuration information indicating basic settings related to the learning model (e.g., information of the sensor 20 used in generating the learning model and information of the sampling interval of the sensor 20).

[0047] The determination unit 240 determines the state of the device 10 using a model for determining the state of the device 10. For example, the determination unit 240 determines the state of the device 10 using the data input from the determination processing unit 250 and the learning model generated by the learning unit 210 and the learning model generation unit 220.

[0048] The judgment processing unit 250 cooperates with the judgment unit 240 to perform judgment processing of the device 10 and write the judgment result to the database of the cloud platform 50. For example, the judgment processing unit 250 inputs the data obtained from the cloud via the interface unit 110 and the cloud communication unit 120 to the judgment unit 240, and uses the judgment unit 240 to perform the state judgment processing of the device 10. In addition, the judgment processing unit 250 writes the judgment result to the database of the cloud platform 50 via the interface unit 110 and the cloud communication unit 120.

[0049] Figure 2 An example of a generation process of a learning model using the interface device 100 of the present embodiment is shown. In addition, before this process, a process of initializing the AI ​​processing unit 200 may be performed. For example, the interface unit 110 may send an initialization instruction to the learning model generation unit 220. Accordingly, the AI ​​processing unit 200 is initialized. In addition, the learning model generation unit 220 may send a SUCCESS instruction indicating that the initialization is successful to the interface unit 110.

[0050] In step 2000, the AI ​​processing unit 200 requests to obtain a list of sensors 20 set in a device 10 managed by any cloud platform 50. For example, the learning model generation unit 220 sends a sensor list acquisition request instruction requesting to obtain a list of sensors 20 set in a device 10 managed by the first cloud platform 50A as a first instruction using a universal format protocol that is not dependent on the cloud platform 50 to the interface unit 110. The interface unit 110 receives the sensor list acquisition request instruction using the universal format protocol as the first instruction. In addition, the interface unit 110 supplies the first instruction received from the learning model generation unit 220 to the first cloud communication unit 120A corresponding to the first cloud platform 50A. Thus, the first instruction received by the interface unit 110 may include a sensor list acquisition request instruction requesting to obtain a list of sensors 20 set in a device 10 managed by the first cloud platform 50A.

[0051] In step 2010, the cloud communication unit 120 converts the first instruction using the universal format protocol supplied in step 2000 into a second instruction using a protocol in a format that is independent of the cloud platform 50. For example, the first cloud communication unit 120A converts the first instruction, which is a sensor list acquisition request instruction for requesting to acquire a list of sensors 20 provided in the device 10 managed by the first cloud platform 50A, into a second instruction using a protocol in a first format (a format corresponding to the REST API) that is independent of the cloud platform 50.

[0052] In step 2020, the cloud communication unit 120 sends the second instruction converted in step 2010 to any cloud platform 50. For example, the first cloud communication unit 120A sends the second instruction converted in step 2010 to the first cloud platform 50A.

[0053] In step 2030, any cloud platform 50 responds to the second instruction received in step 2020. For example, the first cloud platform 50A sends a sensor list acquisition instruction representing a list of sensors 20 provided in the device 10 managed by itself to the first cloud communication unit 120A as a third instruction using the first format protocol based on the second instruction received in step 2020.

[0054] In step 2040, the cloud communication unit 120 converts the third instruction using the inherent format protocol received in step 2030 into a fourth instruction using the universal format protocol. For example, the first cloud communication unit 120A converts the third instruction, which is a sensor list acquisition instruction indicating a list of sensors 20 provided in the device 10 managed by the first cloud platform 50A, and uses the first format protocol dependent on the first cloud platform 50A, into a fourth instruction using the universal format protocol not dependent on the cloud platform 50.

[0055] In step 2050, the cloud communication unit 120 sends the fourth instruction converted in step 2040 to the AI ​​processing unit 200 via the interface unit 110. For example, the first cloud communication unit 120A supplies the fourth instruction converted in step 2040 to the interface unit 110. In addition, the interface unit 110 sends the sensor list acquisition instruction using the universal format protocol to the learning model generation unit 220. Accordingly, the learning model generation unit 220 supplies the fourth instruction received from the interface unit 110 to the learning model management unit 230.

[0056] In step 2060, the AI ​​processing unit 200 stores information related to the list of sensors 20 provided in the device 10 managed by any cloud platform 50. For example, the learning model management unit 230 stores information related to the list of sensors 20 provided in the device 10 managed by the first cloud platform 50A according to the fourth instruction received in step 2050. Thus, in steps 2000 to 2060, the acquisition process of the sensor list using the interface device 100 is performed.

[0057] In step 2100, the AI ​​processing unit 200 requests to obtain the measurement data of the sensor 20 set in the device 10 managed by any cloud platform 50. For example, the learning model generation unit 220 determines the sensor 20 required for the generation of the learning model from the sensor list obtained in step 2050. In addition, the learning model generation unit 220 sends a sensor data acquisition request instruction requesting the acquisition of the measurement data of the determined sensor 20 as a first instruction using a universal format protocol that is not dependent on the cloud platform 50 to the interface unit 110. The interface unit 110 receives the sensor data acquisition request instruction using the universal format protocol as the first instruction. In addition, the interface unit 110 supplies the first instruction received from the learning model generation unit 220 to the first cloud communication unit 120A corresponding to the first cloud platform 50A. Thus, the first instruction received by the interface unit 110 may include a sensor data acquisition request instruction requesting the acquisition of the measurement data of the sensor 20 set in the device 10 managed by the first cloud platform 50A.

[0058] In step 2110, the cloud communication unit 120 converts the first instruction using the universal format protocol supplied in step 2100 into a second instruction using a unique format protocol that is dependent on any cloud platform 50. For example, the first cloud communication unit 120A converts the first instruction, which is a sensor data acquisition request instruction for requesting to acquire measurement data of the sensor 20 provided in the device 10 managed by the first cloud platform 50A, into a second instruction using a first format protocol that is dependent on the first cloud platform 50A. The first instruction uses a universal format protocol that is not dependent on the cloud platform 50.

[0059] In step 2120, the cloud communication unit 120 sends the second instruction converted in step 2110 to any cloud platform 50. For example, the first cloud communication unit 120A sends the second instruction converted in step 2110 to the first cloud platform 50A.

[0060] In step 2130, any cloud platform 50 responds to the second instruction received in step 2120. For example, the first cloud platform 50A sends a sensor data acquisition instruction representing the measurement data of the sensor 20 provided in the device 10 managed by the first cloud platform 50A as a third instruction using the first format protocol to the first cloud communication unit 120A based on the second instruction received in step 2120.

[0061] In step 2140, the cloud communication unit 120 converts the third instruction using the inherent format protocol received in step 2130 into a fourth instruction using the universal format protocol. For example, the first cloud communication unit 120A converts the third instruction, which is a sensor data acquisition instruction indicating measurement data of the sensor 20 provided in the device 10 managed by the first cloud platform 50A, into a fourth instruction using the universal format protocol that is not dependent on the cloud platform 50, and the third instruction uses the first format protocol that is dependent on the first cloud platform 50A.

[0062] In step 2150, the cloud communication unit 120 sends the fourth instruction converted in step 2140 to the AI ​​processing unit 200 via the interface unit 110. For example, the first cloud communication unit 120A supplies the fourth instruction converted in step 2140 to the interface unit 110. In addition, the interface unit 110 sends the sensor data acquisition instruction using the universal format protocol to the learning model generation unit 220. Thus, in steps 2100 to 2150, the sensor data acquisition process using the interface device 100 is performed.

[0063] In step 2200, the AI ​​processing unit 200 calculates the sampling interval. For example, before generating the learning model, the learning model generation unit 220 calculates the sampling interval based on the measurement data acquired in time series in step 2150. For example, the sampling interval in the learning model generation unit 220 can be calculated using the time when the interface unit 110 receives the fourth instruction, in addition to the time when the sensor 20 acquires the data (given to the sensor data).

[0064] In step 2210, the AI ​​processing unit 200 performs preprocessing. For example, before generating the learning model, the learning model generation unit 220 performs preprocessing on the measurement data obtained in step 2150. Such preprocessing may include, for example, interpolation (previous value, average value, median value, mode value, or straight line), waveform compression, moving average, standardization, or normalization.

[0065] In step 2220 and step 2230, the AI ​​processing unit 200 generates a model for determining the state of the device 10 through machine learning. For example, in step 2220, the learning model generation unit 220 inputs the measurement data obtained in step 2150 and preprocessed in step 2210 into the learning unit 210. Accordingly, the learning unit 210 performs machine learning to generate a learning model (AI model) for determining the state of the device 10. And, in step 2230, the learning model generation unit 220 obtains the generated learning model from the learning unit 210. Here, various algorithms can be used for machine learning for generating a learning model. Since this learning process itself is known, its description is omitted here.

[0066] In step 2240, the learning model generation unit 220 supplies the information of the learning model generated in steps 2220 and 2230 together with the configuration information representing the basic settings related to the learning model (for example, information about the sensor 20 (sensor 20 determined in step 2100) used in the generation of the learning model and information about the sampling interval of the sensor 20) to the learning model management unit 230.

[0067] In step 2250, the AI ​​processing unit 200 stores the information of the learning model together with the configuration information. For example, the learning model management unit 230 stores the information of the learning model supplied in step 2240 in a manner corresponding to the information related to the sensor 20 used in generating the learning model and the information related to the sampling interval of the sensor 20. Thus, in steps 2200 to 2250, the generation process of the learning model using the interface device 100 is performed.

[0068] Such sensor data acquisition processing (steps 2100 to 2150) and learning model generation processing (steps 2200 to 2250) may be repeatedly performed according to the number of sensors 20 for generating learning models. In addition, in the above description, as an example, the AI ​​processing unit 200 generates one learning model using measurement data obtained from one sensor 20, but the present invention is not limited thereto. The AI ​​processing unit 200 may also generate one or more learning models using measurement data obtained from a plurality of sensors 20.

[0069] Figure 3 An example of a judgment process using the learning model of the interface device 100 of this embodiment is shown. This process can be performed, for example, Figure 2 The generation process of the learning model shown is then executed.

[0070] In step 3000 and step 3010, the AI ​​processing unit 200 prepares a model for determining the state of the device 10. For example, in step 3000, the determination processing unit 250 requests the learning model management unit 230 for the learning model for state determination. And, in step 3010, the learning model management unit 230 supplies the learning model for state determination to the determination unit 240. In addition, the learning model supplied here can be, for example, Figure 2 It may be generated by the process of

[0071] The processing of steps 3100 to 3150 and the processing of step 3200 are respectively Figure 2 The processing from step 2100 to step 2150 is the same as the processing of step 2210, and the description is omitted here.

[0072] In step 3210 and step 3220, the AI ​​processing unit 200 uses the model to determine the state of the device 10. For example, in step 3210, the judgment processing unit 250 inputs the measurement data obtained in step 3150 and preprocessed in step 3200 in the same manner as in the learning to the judgment unit 240. Based on this, the judgment unit 240 performs a judgment process using the learning model to determine the state of the device 10. And, in step 3220, the judgment processing unit 250 obtains a judgment result of judging the state of the device 10 from the judgment unit 240. Since this judgment process itself is already known, the description is omitted here.

[0073] In step 3300, the AI ​​processing unit 200 requests that the judgment result of the state of the device 10 be written to the database of any cloud platform 50. For example, the judgment processing unit 250 sends a judgment result write request instruction as a first instruction using a universal format protocol that is not dependent on the cloud platform 50 to the interface unit 110, and the judgment result write request instruction requests that the judgment result of the state of the device 10 managed by the first cloud platform 50A be written to the database of the first cloud platform 50A. The interface unit 110 receives the judgment result write request instruction using the universal format protocol as the first instruction. In addition, the interface unit 110 supplies the first instruction received from the judgment processing unit 250 to the first cloud communication unit 120A. Thus, the first instruction received by the interface unit 110 may include a judgment result write request instruction, which requests that the judgment result of the AI ​​processing unit 200 judging the state of the device 10 be written to the database of the first cloud platform 50A.

[0074] In step 3310, the cloud communication unit 120 converts the first instruction using the universal format protocol supplied in step 3300 into a second instruction using a platform that uses an inherent format that depends on any cloud platform 50. For example, the first cloud communication unit 120A converts the first instruction into a second instruction using a first format protocol that depends on the first cloud platform 50A, the first instruction being a judgment result write request instruction that requests writing the judgment result of the state of the device 10 managed by the first cloud platform 50A to the database of the first cloud platform 50A, and the first instruction uses a universal format protocol that does not depend on the cloud platform 50.

[0075] In step 3320, the cloud communication unit 120 sends the second instruction converted in step 3310 to any cloud platform 50. For example, the first cloud communication unit 120A sends the second instruction converted in step 3310 to the first cloud platform 50A.

[0076] In step 3330 , any cloud platform 50 writes the judgment result. For example, the first cloud platform 50A writes the judgment result of the AI ​​processing unit 200 judging the status of the device 10 into its own database according to the second instruction received in step 3320 .

[0077] In step 3340, any cloud platform 50 sends a response to the second instruction to the cloud communication unit 120. For example, the first cloud platform 50A sends a SUCCESS instruction indicating successful writing to the first cloud communication unit 120A as a third instruction using the first format protocol based on successfully writing the judgment result to its own database.

[0078] In step 3350, the cloud communication unit 120 converts the third instruction using the inherent format protocol received in step 3340 into a fourth instruction using the universal format protocol. For example, the first cloud communication unit 120A converts the third instruction into a fourth instruction using the universal format protocol that is not dependent on the cloud platform 50, the third instruction is a SUCCESS instruction indicating that the judgment result is successfully written to the database of the first cloud platform 50A, and the third instruction uses the first format protocol that is dependent on the first cloud platform 50A.

[0079] In step 3360, the cloud communication unit 120 sends the fourth instruction converted in step 3350 to the AI ​​processing unit 200 via the interface unit 110. For example, the first cloud communication unit 120A supplies the fourth instruction converted in step 3350 to the interface unit 110. And, the interface unit 110 sends the SUCCESS instruction using the universal format protocol to the judgment processing unit 250. Thus, in steps 3000 to 3360, the judgment processing using the learning model of the interface device 100 is performed.

[0080] This learning model determination process (steps 3000 to 3360) may be repeatedly performed according to the number of learning models used to determine the state of the device 10. In addition, in the above description, as an example, the AI ​​processing unit 200 uses the measurement data obtained from one sensor 20 to determine the state of the device 10, but it is not limited to this. The AI ​​processing unit 200 may also use the measurement data obtained from multiple sensors 20 to determine the state of the device 10.

[0081] Figure 4 An example of a flow of updating a learning model from the external system 300 using the interface device 100 of the present embodiment is shown. Here, the external system 300 is, for example, a system provided outside the interface device 100 and having a function of generating a customized learning model.

[0082] The processing from step 4000 to step 4050 is similar to that of step 4000 except that the learning model generation unit 220 is changed to the external system 300. Figure 2 The processing from step 2100 to step 2150 is the same, so the description is omitted here.

[0083] In step 4100 , the external system 300 generates a model for determining the state of the device 10 using the measurement data supplied in step 4050 .

[0084] In step 4200, the external system 300 requests to change (e.g., add, update, and delete) at least any one of the information or configuration information of the learning model stored in the internal memory (e.g., the learning model management unit 230) of the cloud platform 50. For example, in step 4200, the external system 300 sends at least any one of a model information change request instruction requesting to change the information of the model managed by the cloud platform 50 or a configuration information change request instruction requesting to change the configuration information to the learning model management unit 230.

[0085] In step 4210, the learning model management unit 230 changes the information managed by itself according to the instruction received in step 4200. For example, the learning model management unit 230 changes (e.g., adds, updates, and deletes) the information of the model managed by itself according to the model information change request instruction received in step 4200. In addition, the learning model management unit 230 changes (e.g., adds, updates, and deletes) the configuration information managed by itself according to the configuration information change request instruction received in step 4200.

[0086] In step 4220, the learning model management unit 230 sends a response to the command received in step 4200 to the external system 300. For example, the learning model management unit 230 sends a SUCCESS command indicating that the information has been successfully changed to the external system 300 based on the successful change of the information managed by the learning model management unit 230.

[0087] Thus, the interface device 100 supports the external system 300 in acquiring measurement data to generate a learning model, and changes the information or configuration information of the model stored in the internal memory of the cloud platform 50 according to the learning model generated by the external system 300. Figure 4 Execute after the process Figure 3 The process uses the learning model generated by the external system 300 to perform state judgment processing.

[0088] Figure 5 Another example of the process of updating the learning model from the external system 300 using the interface device 100 of this embodiment is shown. Figure 4 The processing from step 4000 to step 4100 in is the same, so the description is omitted here. Figure 4 The process shows a situation in which the external system 300 directly requests the learning model management unit 230 to change the information stored in the internal memory of the cloud platform 50. In this process, the external system 300 requests the learning model management unit 230 via the cloud platform 50 to change the information stored in the internal memory of the cloud platform 50.

[0089] In step 5200, the external system 300 requests to change at least any one of the information or configuration information of the learning model stored in the internal memory of any cloud platform 50. For example, in step 5200, the external system 300 sends at least any one of the model information change request instruction requesting to change the information of the model managed by the first cloud platform 50A or the configuration information change request instruction requesting to change the configuration information to the interface unit 110 as a first instruction using a universal format protocol that is not dependent on the cloud platform 50. In addition, the interface unit 110 supplies the first instruction received from the external system 300 to the first cloud communication unit 120A corresponding to the first cloud platform 50A. Thus, the first instruction received by the interface unit 110 may include a model information change request instruction requesting to change the information of the model managed by the first cloud platform 50A. In addition, the first instruction received by the interface unit 110 may include a configuration information change request instruction requesting to change the configuration information managed by the first cloud platform 50A.

[0090] In step 5210, the cloud communication unit 120 converts the first instruction using the universal format protocol supplied in step 5200 into a second instruction using a unique format protocol that is dependent on any cloud platform 50. For example, the first cloud communication unit 120A converts the first instruction into a second instruction using a first format protocol that is dependent on the first cloud platform 50A, the first instruction being at least one of a model information change request instruction requesting a change in information of a model managed by the first cloud platform 50A or a configuration information change request instruction requesting a change in configuration information, and the first instruction using a universal format protocol that is not dependent on the cloud platform 50.

[0091] In step 5220, the cloud communication unit 120 sends the second instruction converted in step 5210 to any cloud platform 50. For example, the first cloud communication unit 120A sends the second instruction converted in step 5210 to the first cloud platform 50A.

[0092] In step 5230, any cloud platform 50 transmits the second instruction received in step 5220 to the learning model management unit 230. For example, the first cloud platform 50A transmits the second instruction received in step 5220 to the learning model management unit 230 attached to itself.

[0093] In step 5240, the learning model management unit 230 changes the information it manages according to the second instruction received in step 5230. Figure 4 The processing of step 4210 in is the same, so the description is omitted here.

[0094] In step 5250, the learning model management unit 230 sends a response to the second instruction received in step 5230 to any cloud platform 50. For example, the learning model management unit 230 sends a SUCCESS instruction indicating that the information has been successfully changed to the first cloud platform 50A based on the successful change of the information managed by itself.

[0095] In step 5260, any cloud platform 50 sends a response to the instruction received in step 5250 to the cloud communication unit 120. For example, the first cloud platform 50A receives the SUCCESS instruction in step 5250 and sends the instruction to the first cloud communication unit 120A as a third instruction using the first format protocol.

[0096] In step 5270, the cloud communication unit 120 converts the third instruction received in step 5260 into a fourth instruction using a universal format protocol. For example, the first cloud communication unit 120A converts the third instruction into a fourth instruction using a universal format protocol that is not dependent on the cloud platform 50, the third instruction is a SUCCESS instruction indicating that the information change is successful, and the third instruction uses a first format protocol that is dependent on the first cloud platform 50A.

[0097] In step 5280, the cloud communication unit 120 sends the fourth instruction converted in step 5270 to the external system 300 via the interface unit 110. For example, the first cloud communication unit 120A supplies the fourth instruction converted in step 5270 to the interface unit 110. And, the interface unit 110 sends a SUCCESS instruction using the universal format protocol to the external system 300.

[0098] Thus, the interface device 100 also provides an interface for changing the information stored in the internal memory of the cloud platform 50 from the external system 300 via the cloud platform. Thus, no matter which cloud platform 50 is used, it is easy to access the information stored in the internal memory of the cloud platform 50 from the external system 300.

[0099] In recent years, the cloud platforms that can be selected have become diversified, and it is not easy to install the functional unit that performs AI processing on multiple cloud platforms. Here, it is preferred to provide a system in which the cloud platform and the functional unit that performs AI processing are integrated, but the actual situation is that there are still users who have already imported the cloud system, and sometimes only newly import the functional unit that performs AI processing in an additional manner. In this case, the system engineer responsible for cloud construction needs to understand the interface data of the already imported platform and the newly imported functional unit that performs AI processing and install them, which makes the construction operation complicated. In addition, installation operation errors are prone to occur during the engineering process.

[0100] In contrast, the interface device 100 of the present embodiment provides an interface between the AI ​​processing unit 200 that performs AI processing and the cloud platform 50. In particular, the interface device 100 of the present embodiment receives a first instruction using a universal format protocol that is not dependent on the cloud platform 50 from the AI ​​processing unit 200. In addition, the interface device 100 converts the first instruction using the universal format protocol into a second instruction using an inherent format protocol that depends on any cloud platform 50, and sends the converted second instruction to any cloud platform 50. Thus, according to the interface device 100 of the present embodiment, the AI ​​processing unit 200 does not need to generate inherent instructions for each cloud platform 50, and the load of the AI ​​processing unit 200 can be reduced. In addition, since the AI ​​processing unit 200 does not depend on various specifications of the cloud platform 50, the impact on the cloud platform 50 can be ignored, so that the AI ​​processing unit 200 can be easily updated. Thus, according to the interface device 100 of the present embodiment, the functional unit that performs AI processing can be easily installed in various cloud environments.

[0101] In addition, the interface device 100 of the present embodiment receives a third instruction using a protocol in a format that is dependent on the cloud platform 50 from the cloud platform 50. Furthermore, the interface device 100 converts the third instruction using the protocol in the format that is dependent on the cloud platform 50 into a fourth instruction using a protocol in a universal format that is not dependent on the cloud platform 50, and sends the converted fourth instruction to the AI ​​processing unit 200. Thus, according to the interface device 100 of the present embodiment, an interface in two directions, from the AI ​​processing unit 200 to the cloud platform 50 and from the cloud platform 50 to the AI ​​processing unit 200, can be provided.

[0102] In addition, the interface device 100 of this embodiment has a plurality of inherent communication units corresponding to the plurality of cloud platforms 50. Thus, according to the interface device 100 of this embodiment, it is possible to provide an interface between the AI ​​processing unit 200 and a plurality of cloud platforms 50 having different specifications. In addition, according to the interface device 100 of this embodiment, it is possible to provide an interface between the AI ​​processing unit 200 and a new cloud platform 50 simply by adding an inherent communication unit.

[0103] In addition, the interface device 100 of this embodiment receives at least any one of a sensor list acquisition request instruction, a sensor data acquisition request instruction, a judgment result write request instruction, a model information change request instruction, or a configuration information change request instruction as a first instruction. Thus, the interface device 100 of this embodiment can relay instructions related to AI processing (learning processing or judgment processing, etc.) for determining the state of the device 10 between the AI ​​processing unit 200 and the cloud platform 50.

[0104] In addition, the interface device 100 of this embodiment includes, for example, an AI processing unit having a learning unit and a determination unit. Therefore, according to the interface device 100 of this embodiment, the interface function and the function of executing AI processing can be configured as an integrated device.

[0105] Various embodiments of the present invention may be described with reference to flow charts and block diagrams, where a module may represent (1) a stage of a process for performing an operation or (2) a portion of a device having the function of performing an operation. Specific stages and portions may be implemented by dedicated circuits, programmable circuits supplied together with computer-readable instructions stored on a computer-readable medium, and / or processors supplied together with computer-readable instructions stored on a computer-readable medium. Dedicated circuits may include digital and / or analog hardware circuits, and may also include integrated circuits (ICs) and / or discrete circuits. Programmable circuits may include reconfigurable hardware circuits, which include logical AND, logical OR, logical XOR, logical NAND, logical NOR and other logical operations, flip-flops, registers, field programmable gate arrays (FPGAs), programmable logic arrays (PLAs) and other memory elements, etc.

[0106] Computer-readable media may include any tangible device capable of storing instructions executed by an appropriate device, and as a result, a computer-readable medium having instructions stored therein includes a product containing instructions that can be executed in order to make a means for performing the operations specified by the flowchart or block diagram. Examples of computer-readable media may include: electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable media may include: floppy (registered trademark) disks, magnetic disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), electrically erasable programmable read-only memories (EEPROM), static random access memories (SRAM), compact disk read-only memories (CD-ROM), digital versatile disks (DVD), Blu-ray (RTM) disks, memory sticks, integrated circuit cards, etc.

[0107] Computer readable instructions include any of source code and object code described by any combination of one or more programming languages ​​including assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or object-oriented programming languages ​​such as Smalltalk (registered trademark), JAVA (registered trademark), C++, and existing procedural programming languages ​​such as the "C" programming language or similar programming languages.

[0108] The computer readable instructions may be provided to a processor or programmable circuit of a general-purpose computer, a special-purpose computer or other programmable data processing device via a local or local area network (LAN), a wide area network (WAN) such as the Internet, and the computer readable instructions are executed to produce a means for performing the operations specified by the flowchart or block diagram. Examples of processors include: computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

[0109] Figure 6 The computer 9900 is an example of a computer that can implement various modes of the present invention in whole or in part. The computer 9900 can perform operations associated with the device of the embodiment of the present invention or functions of one or more parts of the device, or perform the operations or the one or more parts, and / or the computer 9900 can perform processes of the embodiment of the present invention or stages of the processes, through the program installed in the computer 9900. In order to make the computer 9900 perform specific operations associated with some or all of the modules in the flowcharts and block diagrams described in this specification, the CPU 9912 can execute such a program.

[0110] The computer 9900 of the present embodiment includes a CPU 9912, a RAM 9914, a graphic controller 9916, and a display device 9918, which are connected to each other via a main controller 9910. The computer 9900 also includes an input / output unit such as a communication interface 9922, a hard disk drive 9924, a DVD-ROM drive 9926, and an IC card drive, which are connected to the main controller 9910 via an input / output controller 9920. The computer also includes a conventional input / output unit such as a ROM 9930 and a keyboard 9942, which are connected to the input / output controller 9920 via an input / output chip 9940.

[0111] The CPU 9912 controls each unit by operating according to the program stored in the ROM 9930 and the RAM 9914. The graphic controller 9916 acquires image data generated by the CPU 9912 in a frame buffer or the like provided in the RAM 9914 or in itself, and displays the image data on the display device 9918.

[0112] The communication interface 9922 can communicate with other electronic devices via a network. The hard disk drive 9924 stores programs and data used by the CPU 9912 in the computer 9900. The DVD-ROM drive 9926 reads programs or data from the DVD-ROM 9901 and provides the programs or data to the hard disk drive 9924 via the RAM 9914. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0113] The ROM 9930 stores therein a boot program or the like executed by the computer 9900 upon activation and / or a program depending on the hardware of the computer 9900. The input / output chip 9940 may also connect various input / output units to the input / output controller 9920 via a parallel port, a serial port, a keyboard port, a mouse port, or the like.

[0114] The program is provided by a computer-readable medium such as DVD-ROM 9901 or an IC card. The program is read from the computer-readable medium, and installed in the hard disk drive 9924, RAM 9914, or ROM 9930, which is also an example of a computer-readable medium, and executed by the CPU 9912. The information processing described in these programs is read to the computer 9900, thereby bringing about cooperation between the program and the various types of hardware resources described above. The device or method can be configured by realizing the operation or processing of information with the use of the computer 9900.

[0115] For example, in the case of performing communication between the computer 9900 and an external device, the CPU 9912 can execute a communication program loaded in the RAM 9914, and instruct the communication interface 9922 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 9912, the communication interface 9922 reads transmission data stored in a transmission buffer processing area provided in a recording medium such as the RAM 9914, the hard disk drive 9924, the DVD-ROM 9901, or an IC card, and transmits the read transmission data to the network, or writes reception data received from the network to a reception buffer processing area provided on the recording medium, etc.

[0116] Furthermore, the CPU 9912 can read all or a necessary part of a file or database stored in an external recording medium such as the hard disk drive 9924, the DVD-ROM drive 9926 (DVD-ROM 9901), or an IC card into the RAM 9914, and perform various types of processing on the data in the RAM 9914. Then, the CPU 9912 writes the processed data back to the external recording medium.

[0117] Various types of information such as various types of programs, data, tables, and databases can be stored in a recording medium and subjected to information processing. The CPU 9912 performs various types of processing described in various places of the present disclosure on the data read from the RAM 9914 and writes the results back to the RAM 9914. The various types of processing include various types of operations specified by the instruction sequence of the program, information processing, conditional judgment, conditional branching, unconditional branching, information retrieval / replacement, etc. In addition, the CPU 9912 can retrieve information in files, databases, etc. in the recording medium. For example, in the case where multiple entries having attribute values ​​of a first attribute respectively associated with attribute values ​​of a second attribute are stored in the recording medium, the CPU 9912 can retrieve an entry consistent with the condition specifying the attribute value of the first attribute from the multiple entries, and read the attribute value of the second attribute stored in the entry, thereby obtaining the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0118] The program or software module described above may be stored in a computer-readable medium on or near the computer 9900. In addition, a recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can be used as a computer-readable medium, thereby providing the program to the computer 9900 via the network.

[0119] The present invention has been described above using the embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments. It is obvious to those skilled in the art that various changes or improvements can be made to the above embodiments. According to the description of the claims, the method of making such changes or improvements can also be included in the technical scope of the present invention.

[0120] The execution order of each process such as actions, processes, steps and stages in the apparatus, system, program and method shown in the claims, specifications and drawings is not specifically indicated as "earlier", "before", etc. In addition, it should be noted that as long as the output of the previous process is not used in the subsequent process, it can be implemented in any order. Even if the action flow in the claims, specifications and drawings is described using "first," "next," etc. for the sake of convenience, it does not mean that it must be implemented in this order.

Claims

1. An interface device, characterized in that include: an interface unit that receives a first instruction using a universal format protocol that is independent of a cloud platform that manages the device from an AI processing unit that performs at least one of a process of generating a model for determining a state of a device through machine learning or a process of determining the state of the device using the model; as well as a cloud communication unit, converting the first instruction into a second instruction using a format protocol dependent on the cloud platform, and sending the second instruction to the cloud platform, The cloud communication unit has: The inherent first cloud communication unit converts the first instruction for the first cloud platform into the second instruction using a first format protocol corresponding to the first interface provided by the first cloud platform, and sends the second instruction using the first format protocol to the first cloud platform; as well as The inherent second cloud communication unit converts the first instruction for the second cloud platform into the second instruction using the second format protocol corresponding to the second interface provided by the second cloud platform, and sends the second instruction using the second format protocol to the second cloud platform.

2. The interface device according to claim 1, characterized in that: The cloud communication unit receives a third instruction using a format protocol dependent on the cloud platform from the cloud platform, and converts the third instruction into a fourth instruction using the universal format protocol, The interface unit sends the fourth instruction to the AI ​​processing unit.

3. The interface device according to claim 1 or 2, characterized in that: The first instruction includes a sensor list acquisition request instruction for requesting to acquire a list of sensors provided in the device managed by the cloud platform.

4. The interface device according to claim 1 or 2, characterized in that: The first instruction includes a sensor data acquisition request instruction for requesting acquisition of measurement data of a sensor provided in the device managed by the cloud platform.

5. The interface device according to claim 1 or 2, characterized in that: The first instruction includes a judgment result write request instruction for requesting to write a judgment result of the status of the device into a database of the cloud platform.

6. The interface device according to claim 1 or 2, characterized in that: The first instruction includes a model information change request instruction for requesting to change the information of the model managed by the cloud platform.

7. The interface device according to claim 1 or 2, characterized in that: The first instruction includes a configuration information change request instruction for requesting to change the configuration information managed by the cloud platform.

8. The interface device according to claim 1 or 2, characterized in that: Also included is the AI ​​processing unit.

9. The interface device according to claim 8, characterized in that: The AI ​​processing unit includes a learning unit that generates the model through machine learning.

10. The interface device according to claim 8, characterized in that: The AI ​​processing unit includes a determination unit that determines the state of the device using the model.

11. An interface method, characterized in that include: receiving, from an AI processing unit that performs at least one of a process of generating a model for determining a state of a device through machine learning or a process of determining the state of the device using the model, a first instruction using a universal format protocol that is independent of a cloud platform that manages the device; as well as converting the first instruction into a second instruction using a format protocol dependent on the cloud platform, and sending the second instruction to the cloud platform, Converting into the second instruction and sending the second instruction to the cloud platform comprises the following steps: The inherent first cloud communication unit converts the first instruction for the first cloud platform into the second instruction using a first format protocol corresponding to the first interface provided by the first cloud platform, and sends the second instruction using the first format protocol to the first cloud platform; as well as The inherent second cloud communication unit converts the first instruction for the second cloud platform into the second instruction using the second format protocol corresponding to the second interface provided by the second cloud platform, and sends the second instruction using the second format protocol to the second cloud platform.

12. A recording medium having an interface program recorded thereon, characterized in that: The computer functions as an interface unit and a cloud communication unit by executing the interface program. The interface unit receives a first instruction using a universal format protocol that is independent of a cloud platform that manages the device from an AI processing unit that performs at least one of a process of generating a model for determining the state of the device through machine learning or a process of determining the state of the device using the model, The cloud communication unit converts the first instruction into a second instruction using a format protocol that depends on the cloud platform, and sends the second instruction to the cloud platform, The cloud communication unit has: The inherent first cloud communication unit converts the first instruction for the first cloud platform into the second instruction using a first format protocol corresponding to the first interface provided by the first cloud platform, and sends the second instruction using the first format protocol to the first cloud platform; as well as The inherent second cloud communication unit converts the first instruction for the second cloud platform into the second instruction using the second format protocol corresponding to the second interface provided by the second cloud platform, and sends the second instruction using the second format protocol to the second cloud platform.

Images