The present disclosure relates to a controller, a terminal device, a management system, and a management method.
For example, Japanese Patent No. 5510926 (PTL 1) discloses a detection method for detecting an abnormality in a blade bearing of a wind turbine generator.
PTL 1: Japanese Patent No. 5510926
The above-mentioned detection method took no consideration of a technique for allowing a maintenance person or the like to recognize maintenance information for eliminating an abnormality in the wind turbine generator.
The present disclosure has been made to solve the above-described problem,
and an object thereof is to provide a technique for allowing a maintenance person or the like to recognize maintenance information for eliminating an abnormality in a wind turbine generator.
A management system according to the present disclosure includes a detection unit, a controller, and a terminal device. The detection unit acquires detection data of a wind turbine generator. The controller generates maintenance information related to maintenance for the wind turbine generator based on the detection data. The controller is configured to transmit the maintenance information to a first terminal device.
A controller according to the present disclosure includes an input interface, a processor, and an output interface. The input interface receives an input of detection data of a wind turbine generator from a detection unit that acquires the detection data. The processor generates maintenance information related to maintenance for the wind turbine generator based on the detection data. The output interface transmits the maintenance information to a first terminal device.
A management method according to the present disclosure includes: receiving an input of detection data of a wind turbine generator from a detection unit that acquires the detection data; generating maintenance information related to maintenance for the wind turbine generator based on the detection data; and transmitting the maintenance information to a first terminal device.
The present disclosure allows a maintenance person or the like to recognize maintenance information for eliminating an abnormality in a wind turbine generator.
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same or corresponding portions are denoted by the same reference characters, and the description thereof will not be repeated.
Wind turbine generator 20 serves to receive wind power to generate electric power. Wind turbine generator 20 includes a bearing, a power generator, and the like. Detection unit 45 is installed in association with wind turbine generator 20. Detection unit 45 monitors a component of a corresponding wind turbine generator 20 to acquire detection data of the component. The detection data is used for detecting an abnormality in wind turbine generator 20. The component is also referred to as a “detection target component”.
In the example in
Note that the detection data may be other data instead of the data detected by each sensor. Detection unit 45 may include, for example, a camera, and the detection data may be an image captured by the camera.
Controller 100 generates maintenance information based on the detection data. In this case, the maintenance information is related to maintenance of a part (a component) having an abnormality or an abnormality sign in wind turbine generator 20. The information indicating an abnormality sign in wind turbine generator 20 indicates that an abnormality does not occur at present but will occur in the future. In the present embodiment, controller 100 generates maintenance information with the use of artificial intelligence (AI).
Maintenance terminal 72 is a terminal owned by a maintenance service provider or a maintenance team who performs maintenance for wind turbine generator 20. Maintenance terminal 72 is involved with maintenance for wind turbine generator 20. When controller 100 detects an abnormality in wind turbine generator 20, a person in charge of maintenance may perform maintenance for eliminating the abnormality. Thus, controller 100 requests the maintenance service provider or the maintenance team to perform maintenance for wind turbine generator 20. Maintenance terminal 72 serves to receive a request for maintenance. Specifically, controller 100 transmits maintenance information to maintenance terminal 72. A display device of maintenance terminal 72 shows a maintenance image based on the maintenance information. The maintenance image will be described later. A person who performs maintenance for wind turbine generator 20 is also referred to as a “maintenance person”.
Maintenance terminal 72 is a portable terminal that can be carried by a maintenance person. The portable terminal is, for example, a smartphone, a tablet, or the like. Maintenance terminal 72 may be, for example, a stationary terminal (for example, a desktop personal computer). Further, maintenance terminal 72 may be a dedicated computer terminal. The display device of maintenance terminal 72 shows various images as described later.
User terminal 50 is a terminal device owned by a user A. The “user” typically refers to a person who owns wind turbine generator 20. User terminal 50 is typically a portable terminal that can be carried by user A. Further, user terminal 50 may be a dedicated computer terminal.
As shown in
Computing device 11 is a computing entity that executes various programs to perform various processes such as an estimation process and a learning process for an estimation model 121, and is one example of a computer. Computing device 11 is constituted, for example, of a central processing unit (CPU), a field-programmable gate array (FPGA), a graphics processing unit (GPU), and the like. Note that computing device 11 may be constituted of at least one of the CPU, the FPGA, and the GPU, or may be constituted of the CPU and the FPGA, the FPGA and the GPU, the CPU and the GPU, or all of the CPU, the FPGA, and the GPU. Further, computing device 11 may be constituted of processing circuitry.
Storage device 12 includes a volatile storage area (for example, a working area) that temporarily stores a program code, a work memory, and the like when computing device 11 executes an arbitrary program. For example, storage device 12 is constituted of a volatile memory device such as a dynamic random access memory (DRAM) or a static random access memory (SRAM). Further, storage device 12 includes a non-volatile storage area. For example, storage device 12 is constituted of a non-volatile memory device such as a hard disk or a solid state drive (SSD).
Although the present embodiment describes an example in which the volatile storage area and the non-volatile storage area are included in the same storage device 12, the volatile storage area and the non-volatile storage area may be included in different storage devices. For example, computing device 11 may include a volatile storage area, and storage device 12 may include a non-volatile storage area. Controller 100 may include a microcomputer including computing device 11 and storage device 12.
Storage device 12 stores estimation model 121 and an estimation program 122. Estimation model 121 includes a neural network and parameters used in a process in the neural network. The parameters include a weighting coefficient used for calculation by the neural network and a determination value used for determination of the estimation.
Estimation model 121 includes a program at least capable of machine learning
and performs machine learning based on learning data (training data) so that estimation model 121 is optimized (adjusted). The learning data includes input data and ground truth data. The input data includes detection data and a maintenance schedule. As described above, the detection data is to be transmitted from detection unit 45. The maintenance schedule is set by the maintenance person or a user of wind turbine generator 20. For example, in general, a higher temperature in the summer season leads to an increased maintenance burden on the maintenance person. Thus, the maintenance schedule is set to be a time period other than summer by the user or the like. The maintenance schedule is stored, for example, in the RAM of controller 100.
The ground truth data includes maintenance information, which will be described later. Upon receipt of input data, estimation model 121 estimates maintenance information by a neural network 1211 based on the input data. Then, estimation model 121 compares the maintenance information estimated by estimation model 121 itself with the maintenance information included in the ground truth data. Based on this comparison, if the degree of coincidence of information is within a prescribed range, estimation model 121 does not update a parameter 1212, but if the degree of coincidence is out of the prescribed range, estimation model 121 updates parameter 1212 such that the degree of coincidence is within the prescribed range.
Thus, estimation model 121 optimizes parameter 1212. In this way, estimation model 121 performs machine learning by optimizing parameter 1212 with the use of learning data including the input data and the ground truth data.
The process of learning estimation model 121 as described above is also referred to as a “learning process”. Estimation model 121 optimized by the learning process is also particularly referred to as a “learned model”. In other words, in the present embodiment, estimation model 121 before learning and estimation model 121 after learning are collectively referred to as an “estimation model”, and particularly, estimation model 121 after learning is also referred to as a “learned model”.
Estimation program 122 is a program for computing device 11 to perform the estimation process and the learning process. Medium reading device 13 receives a storage medium such as a removable disk 130 and acquires data stored in removable disk 130.
Communication device 14 transmits and receives data to and from another device (user terminal 50 or maintenance terminal 72) via network NW.
Input unit 1101 corresponds to communication device 14 and receives input data. As described above, the input data includes the detection data and the maintenance schedule.
Estimation unit 1102 estimates maintenance information based on the input data input through input unit 1101 and estimation model 121 including neural network 1211. Storage unit 1103 corresponds to storage device 12, and stores estimation model 121 including neural network 1211 and parameter 1212.
Output unit 1104 corresponds to communication device 14, and externally outputs the maintenance information estimated by estimation unit 1102 and the information indicating a terminal device as a target for transmission.
The following describes maintenance information. The maintenance information is optimal information for the maintenance person to perform maintenance for wind turbine generator 20. The maintenance information includes an abnormal part, a maintenance type, a maintenance part, a tool to be used, a used component, and a maintenance timing.
The abnormal part is information indicating a part having an abnormality in wind turbine generator 20. The abnormal part is, for example, a bearing. The maintenance type is information indicating a type of maintenance for wind turbine generator 20. The type of maintenance includes, for example, at least one of a repair of a component of wind turbine generator 20, a position adjustment of the component, and an inspection of the component. The component of wind turbine generator 20 is, for example, a bearing. The position adjustment of the component of wind turbine generator 20 includes, for example, a position adjustment of a main shaft of a bearing of wind turbine generator 20. The inspection of wind turbine generator 20 includes, for example, an inspection of components contained in lubricating oil of wind turbine generator 20.
The maintenance part is information indicating a part where maintenance is performed in wind turbine generator 20. The maintenance part is a part indicating an abnormality or an abnormality sign. The tool to be used is a tool necessary for maintenance of wind turbine generator 20. The tool includes, for example, a stethoscope for listening to abnormal sounds of a bearing.
The used component is, for example, a component that needs to be replaced since it has an abnormality. The used component includes, for example, a speed-up gear, a power generator, a bearing, and the like. The maintenance timing indicates a timing suitable for performing maintenance. The maintenance timing is, for example, winter and the like.
The information indicating the terminal device as a target for transmission indicates the terminal device to which the maintenance information is to be transmitted. For example, in the case where the abnormality in wind turbine generator 20 is severe, the maintenance is performed preferably by the maintenance person. On the other hand, in the case where the abnormality in wind turbine generator 20 is minor, user A does not need to request the maintenance person who requires a maintenance fee for maintenance. Thus, in the case where the abnormality in wind turbine generator 20 is minor, the maintenance information is transmitted to user terminal 50. User terminal 50 displays the maintenance information to allow the user to recognize the maintenance information.
In the present embodiment, the abnormality not requiring a maintenance person (a minor abnormality) is also referred to as a “first abnormality”. Further, the abnormality requiring a maintenance person (a severe abnormality) is also referred to as a “second abnormality”.
As described above, based on the wind turbine generator ID, controller 100 can specify the installation position of wind turbine generator 20 indicated by this wind turbine generator ID. Controller 100 transmits the installation position together with the maintenance information to the terminal device as a target for transmission.
Imaging unit 81 captures an image of wind turbine generator 20 or the like to generate imaging data. The imaging data is output to display control unit 82. Display control unit 82 causes display unit 83 to show an image based on the imaging data.
Further, communication unit 84 receives information (for example, maintenance information) from another device (for example, controller 100). The maintenance information is output to display control unit 82. Display control unit 82 causes display unit 83 to show an image based on the maintenance information. Note that user terminal 50 may have the functions shown in
Storage unit 85 stores a maintenance part. The maintenance part is included in the maintenance information acquired by communication unit 84 (see
Then, a display image displayed on maintenance terminal 72 or user terminal 50 will be hereinafter described. The following mainly describes a display image on maintenance terminal 72. Upon receipt of the maintenance information, maintenance terminal 72 gives a notification about the maintenance information. In this case, “giving a notification” is a process for causing the user to recognize information. Also, “giving a notification” includes a process of displaying the information and a process of outputting sound indicating the information. The present embodiment describes a configuration in which “giving a notification” is a process of displaying the information.
The following describes a display image displayed on display unit 83 of the maintenance terminal.
Image 201 shows a component having an abnormality (hereinafter also referred to as an “abnormal component”). The maintenance person visually checks image 201 and thereby can recognize that “an abnormality has occurred in wind turbine generator 20 and an abnormal component”. In the example in
Image 202 shows a maintenance type. The maintenance person visually checks image 202 and thereby can recognize the maintenance type. In the example in
Image 203 shows a maintenance part. The maintenance person visually checks image 203 and thereby can recognize the maintenance part. In the example in
Image 204 shows a tool to be used. The maintenance person visually checks image 204 and thereby can recognize the tool to be used. In the example in
Image 205 shows a component to be used. The maintenance person visually checks image 205 and thereby can recognize a component to be used. The component to be used is, for example, a replacement component for an abnormal component. In the example in
Image 206 shows a maintenance timing. The maintenance person visually checks image 206 and thereby can recognize the maintenance timing. In the example in
Image 207 shows a maintenance place. The maintenance person visually checks image 207 and thereby can recognize the maintenance place. In the example in
When the maintenance timing arrives, for example, the maintenance person carrying maintenance terminal 72 that is a portable terminal visits the maintenance site (i.e., the position of wind turbine generator 20 as a target to receive maintenance). Then, the maintenance person captures an image of maintenance terminal 72 and also 15 captures an image of the vicinity of an abnormal part in wind turbine generator 20. The captured image may be a still image or a moving image.
As described with reference to
Then, display control unit 82 displays the captured image of the maintenance part in wind turbine generator 20 in a manner different from that of display of the 25 captured images of other parts in wind turbine generator 20.
First, in step S2, controller 100 acquires detection data from detection unit 45. Then, in step S3, controller 100 generates maintenance information based on the detection data. Then, in step S4, controller 100 transmits the maintenance information to maintenance terminal 72. In step S6, maintenance terminal 72 displays the maintenance information (see
As shown in
As shown in
Further, as shown in
Further, as shown in
Further, as shown in
As shown in
Further, as shown in
As shown in
As shown in
Further, as shown in
In the configuration described in the above embodiment, controller 100 generates the maintenance information. However, another device may have the function of generating the maintenance information. For example, another device may be maintenance terminal 72. For example, display control unit 82 shown in
(1) In the above description of the example in
(2) In the configuration described in the above embodiment, controller 100
generates the maintenance information mainly with the use of AI. However, controller 100 may employ another method to generate at least part of the maintenance information. For example, controller 100, a designer or the like may construct a database in which the detection data is associated with the maintenance information, and also, controller 100 may use the database to generate the maintenance information.
(Clause 1) A management system according to the present disclosure includes: a detection unit that acquires detection data of a wind turbine generator; a controller; and a first terminal device. The controller generates maintenance information related to maintenance for the wind turbine generator based on the detection data. The controller is configured to transmit the maintenance information to the first terminal device.
(Clause 2) In the management system according to Clause 1, the controller is configured to specify a maintenance type of the wind turbine generator based on the detection data. The maintenance information includes information indicating the maintenance type.
(Clause 3) In the management system according to Clause 2, the maintenance type includes at least one of a repair of a component of the wind turbine generator, a position adjustment of the component, and an inspection of the component.
(Clause 4) In the management system according to any one of Clauses 1 to 3, the controller is configured to specify a maintenance part of the wind turbine generator based on the detection data. The maintenance information includes information indicating the maintenance part.
(Clause 5) In the management system according to any one of Clauses 1 to 4,
the controller is configured to specify a tool and a component necessary for maintenance for the wind turbine generator based on the detection data. The maintenance information includes information indicating the tool and the component.
(Clause 6) In the management system according to any one of Clauses 1 to 5, the controller is configured to specify a maintenance timing for the wind turbine generator based on the detection data. The maintenance information includes information indicating the maintenance timing.
(Clause 7) In the management system according to any one of Clauses 1 to 6, the controller is configured to specify an abnormal part in the wind turbine generator based on the detection data. The maintenance information includes information indicating the abnormal part.
(Clause 8) In the management system according to any one of Clauses 1 to 7, the management system further includes a second terminal device. The controller is configured to specify a degree of abnormality in the wind turbine generator based on the detection data, and transmit the maintenance information to one of the first terminal device and the second terminal device according to the degree of abnormality.
(Clause 9) In the management system according to any one of Clauses 1 to 8, the first terminal device is a portable terminal device.
(Clause 10) In the management system according to any one of Clauses 1 to 9,
the first terminal device includes: an imaging device that captures an image of the wind turbine generator; a display device that shows a captured image captured by the imaging device; and a display controller. The display controller displays the captured image of a maintenance part in the wind turbine generator in a manner different from a manner of display of the captured image of each of other parts in the wind turbine generator.
(Clause 11) A controller according to the present disclosure includes an input interface, a processor, and an output interface. The input interface receives an input of detection data of a wind turbine generator from a detection unit that acquires the detection data. The processor generates maintenance information related to maintenance for the wind turbine generator based on the detection data. The output interface transmits the maintenance information to a first terminal device.
(Clause 12) A terminal device according to the present disclosure includes an input interface, a processor, and a display. The input interface receives an input of detection data of a wind turbine generator from a detection unit that acquires the detection data. The processor generates maintenance information related to maintenance for the wind turbine generator based on the detection data. The display shows an image based on the maintenance information.
(Clause 13) A management method according to the present disclosure includes: receiving an input of detection data of a wind turbine generator from a detection unit that acquires the detection data; generating maintenance information related to maintenance for the wind turbine generator based on the detection data; and transmitting the maintenance information to a first terminal device.
It should be understood that the embodiments disclosed herein are illustrative and non-restrictive in every respect. The scope of the present invention is defined by the terms of the claims, rather than the above description of the embodiments, and is intended to include any modifications within the meaning and scope equivalent to the terms of the claims.
10 management system, 11 computing device, 12 storage device, 14 communication device, 20 wind turbine generator, 30 collection device, 45 detection unit, 50 user terminal, 72 maintenance terminal, 81 imaging unit, 82 display control unit, 83 display unit, 84 communication unit, 85 storage unit, 100 controller, 121 estimation model, 122 estimation program, 130 removable disk, 201, 202, 203, 204, 205, 206, 207 image, 211 maintenance image, 212 captured image, 832 flat plate portion, 1101 input unit, 1102 estimation unit, 1104 output unit, 1211 neural network.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2022-097400 | Jun 2022 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2023/022048 | 6/14/2023 | WO |