JP2018081588A - Specifying device and specifying method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the cost of equipment used for detecting abnormalities in parts.SOLUTION: A specifying device 100 measures device vibration information using a measurement part. The specifying device 100 compares first vibration information, measured by the measurement part under the condition that multiple parts all operate normally, with second vibration information, measured by the measurement part under the condition that it is uncertain whether multiple parts are operating normally. When the first vibration information and the second vibration information differ, the second vibration information and cause information, which may indicate which of the parts causes abnormalities, are registered to a table in association with each other. When the third vibration information measured by the measurement part and the first vibration information differ, the specifying device 100 specifies cause information of abnormalities on the basis of the third vibration information and the second vibration information in the table.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a specific device and the like.

  In a factory, a desired product is produced by processing various materials using a machine. Such a machine is composed of various parts, and if any of the parts deteriorates with age, the machine operates abnormally and cannot properly produce a product. For this reason, factory workers regularly inspect the machine, determine whether or not the parts have deteriorated over time, and replace the parts as necessary.

  As described above, since the burden on the worker is large when the worker inspects the machine, there is a conventional technique for detecting an abnormality of a component using a sensor. In this prior art, a dedicated sensor is attached to each part included in the machine, the vibration of each part is observed, and if an abnormality is found in the vibration of the part, information on the corresponding part is notified.

JP-A-6-66241 JP-A-5-296888 JP 2013-200144 A

  However, the above-described conventional technique has a problem that the cost of equipment for detecting an abnormality of a component is increased.

  In the prior art, it is assumed that one sensor is installed for each component and an abnormality is detected. However, although the components are cheap, the cost is poor because the sensors are expensive. On the other hand, if an attempt is made to detect an abnormality of a component by simply reducing the number of sensors, vibrations of a plurality of components are mixed in the sensor, making it difficult to identify the component in which an abnormality has occurred. Such a problem is not limited to a machine that produces a product, but also occurs in other machines that are operated by a plurality of parts.

  In one aspect, an object of the present invention is to provide a specifying device and a specifying method that can reduce the cost of equipment for detecting an abnormality of a component.

  In the first plan, the specifying device includes a measurement unit, a learning unit, and a specifying unit. The measurement unit is installed in an apparatus having a plurality of parts that are individually operated, and measures vibration information of the apparatus. The learning unit measured the first vibration information measured by the measurement unit under the condition that all of the multiple parts are operating normally, and the measurement unit under the condition where it is unknown whether the multiple parts are operating normally. When the first vibration information and the second vibration information are different from each other by comparing the second vibration information, the second vibration information is associated with cause information indicating which component is the cause of the abnormality. Register in the table. The identification unit identifies the cause information of the abnormality based on the third vibration information and the second vibration information of the table when the third vibration information measured by the measurement unit is different from the first vibration information. To do.

  The cost of equipment for detecting an abnormality of a component can be reduced.

FIG. 1 is a diagram illustrating an example of a system according to the present embodiment. FIG. 2 is a functional block diagram showing the configuration of the specific device. FIG. 3 is a diagram illustrating an example of a data structure of the normal vibration table. FIG. 4 is a diagram illustrating an example of the first vibration information during operation start. FIG. 5 is a diagram illustrating an example of the first amplitude information during operation. FIG. 6 is a diagram illustrating an example of the first amplitude information during operation stop. FIG. 7 is a diagram illustrating an example of the data structure of the learning table. FIG. 8 is a diagram for explaining processing during the start of operation of the learning unit. FIG. 9 is a diagram for explaining processing during operation of the learning unit. FIG. 10 is a diagram for explaining a process during the operation stop of the learning unit. FIG. 11 is a flowchart illustrating the processing procedure of the specific device in the learning phase. FIG. 12 is a flowchart showing the processing procedure of the specifying unit in the specific phase.

  Hereinafter, embodiments of a specific device and a specific method disclosed in the present application will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

  FIG. 1 is a diagram illustrating an example of a system according to the present embodiment. As shown in FIG. 1, the system includes a machine 40, a sensor 50, and a specific device 100.

  The machine 40 includes a plurality of parts 10a, 10b, 10c, 10d, and 10e, and produces a product 6 by processing the material 5, for example. The components 10a to 10e operate at a timing specific to the components 10a to 10e when the machine 40 starts to be activated. The components 10a to 10e may be all different components, or some of the components may be the same component.

  The sensor 50 is a measuring device that is installed at a predetermined position of the machine 40 and measures vibration information of the entire machine 40. The sensor 50 outputs the measured vibration information to the specific device 100. The sensor 50 is an example of a measurement unit.

  The identification device 100 is a device that acquires vibration information from the sensor 50 and identifies a component that causes an abnormality. For example, the specific device 100 executes processing of a learning phase and a specific phase. The identification device 100 repeatedly executes the learning phase and the identification phase. Further, the identification device 100 may perform the learning phase and the identification phase at the same time. In other words, during the operation of the machine 40, while performing relearning in the learning phase, using the result of relearning, the specific phase is executed to identify the component that causes the abnormality.

  Processing of the learning phase by the specific device 100 will be described. First, the specific apparatus 100 stores the vibration information at the normal time measured by the sensor 50 under the condition that all the components 10a to 10e operate normally. For example, the vibration information at normal time is vibration information measured by the sensor 50 immediately after the maintenance of the machine 40 is completed.

  Subsequently, the specific device 100 compares the vibration information measured by the sensor 50 with the vibration information at the normal time under conditions where it is unknown whether the components 10a to 10e are operating normally. It is determined whether or not they match.

  The identification device 100 outputs a warning when the pieces of vibration information do not match, and receives input of cause information indicating which component is the cause of the abnormality from the outside. In the following description, the vibration information at the normal time is appropriately expressed as first vibration information, and the vibration information that does not coincide with the first vibration information is described as second vibration information. The identification device 100 registers the second vibration information and the abnormality cause information in the learning table in association with each other.

  For example, when an operator finds a warning from the specifying device 100, the worker performs maintenance of the machine 40 and specifies a component that causes an abnormality. The worker inputs information (cause information) of the part that causes the specified abnormality to the specific device 100.

  The identification device 100 repeatedly executes the above process in the learning phase. In addition, when the first vibration information matches the vibration information measured by the sensor 50, the specifying apparatus 100 skips registration with respect to the learning table.

  Furthermore, the factory worker may input the second vibration information to the specific device 100 to learn by referring to the daily report of the factory and inputting it to the specific device 100. The daily report includes the vibration information measured by the sensor 50. When the worker refers to the daily report and finds a singular point in the vibration information, the vibration information of the section including the singular point is obtained as the second vibration information. And input to the specifying device 100.

  A specific phase process by the specific device 100 will be described. The identification device 100 acquires vibration information from the sensor 50 when the machine 40 starts operation in the identification phase. The vibration information in the specific phase is appropriately described as third vibration information. The identification device 100 acquires the third vibration information from the sensor 50 and determines whether or not it matches the first vibration information. The identification device 100 compares the third vibration information and the second vibration information in the learning table when the first vibration information and the third vibration information do not match.

  The identifying device 100 identifies cause information associated with the second vibration information that matches the third vibration information in the learning table. The identification device 100 outputs the identified cause information.

  As described above, in the learning phase, the identification device 100 learns the second vibration information and the cause information at the time of occurrence of the abnormality in the learning table, and uses the learning table so that the single sensor 50 uses the learning table. , Abnormal parts can be identified. For this reason, since the number of sensors for specifying an abnormal part can be reduced, the cost of equipment for detecting an abnormality of the part can be reduced.

  Next, an example of the configuration of the specific device 100 illustrated in FIG. 1 will be described. FIG. 2 is a functional block diagram showing the configuration of the specific device. As shown in FIG. 2, the specifying device 100 is connected to the sensor 50. The identification device 100 includes an input unit 110, a display unit 120, a storage unit 130, and a control unit 140.

  The input unit 110 is an input device for an operator to input various types of information such as cause information to the specific device 100. For example, the input unit 110 corresponds to a keyboard, a mouse, a touch panel, or the like.

  The display unit 120 is a display device that displays information output from the control unit 140. The display unit 120 corresponds to a liquid crystal display, a touch panel, or the like.

  The storage unit 130 includes a normal vibration table 130a and a learning table 130b. The storage unit 130 corresponds to a semiconductor memory device such as a random access memory (RAM), a read only memory (ROM), and a flash memory, and a storage device such as a hard disk drive (HDD).

  The normal vibration table 130a is a table that stores normal vibration information (first vibration information) measured by the sensor 50 under the condition that all the components 10a to 10e operate normally. FIG. 3 is a diagram illustrating an example of a data structure of the normal vibration table. As shown in FIG. 3, the normal vibration table 130a associates the condition with the first vibration information.

  The condition is classified into one of operation start, operation, and operation stop. During operation start, it corresponds to the time from the time when any one of the components 10a to 10e of the machine 40 starts operation until a predetermined time later. During operation, it corresponds to the time when all the operation states of the parts 10a to 10e of the machine 40 are in operation. For example, during operation, it may be a preset time from the first time to the second time. During operation stop, when the time before the predetermined time is the third time from the fourth time when all of the parts 10a to 10e of the machine 40 are stopped, this corresponds to the time from the third time to the fourth time.

  The first vibration information measured during the operation start, the first vibration information measured during the operation, and the first vibration information measured during the operation stop are registered in advance in the first vibration information of the normal vibration table 130a. The

  FIG. 4 is a diagram illustrating an example of the first vibration information during operation start. In FIG. 4, the horizontal axis is an axis indicating time, and the vertical axis is an axis indicating the amplitude of the first vibration information. FIG. 5 is a diagram illustrating an example of the first amplitude information during operation. In FIG. 5, the horizontal axis is an axis indicating time, and the vertical axis is an axis indicating the amplitude of the first vibration information. FIG. 6 is a diagram illustrating an example of the first amplitude information during operation stop. In FIG. 6, the horizontal axis is an axis indicating time, and the vertical axis is an axis indicating the amplitude of the first vibration information.

  The learning table 130b is a table that associates the second vibration information with the cause information in the learning phase as described above. FIG. 7 is a diagram illustrating an example of the data structure of the learning table. As shown in FIG. 7, this learning table 130b associates conditions, second vibration information, and cause information. The description regarding the conditions is the same as the description regarding the conditions of the normal vibration table 130a.

  The second vibration information is the second vibration information when the sensor 50 detects vibration information different from the first vibration information, and the cause information input from the input unit 110 when the second vibration information is detected. It is associated. Further, the relationship between the second vibration information and the cause information is classified and stored in the conditions “being started”, “being operated”, and “being stopped”.

  Returning to the description of FIG. The control unit 140 includes a learning unit 140a and a specifying unit 140b. The control unit 140 can be realized by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. It can also be realized by hard wired logic such as ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

  The learning unit 140a is a processing unit that repeatedly executes a process of storing the second vibration information and the cause information in association with each other in the learning table 130b in the learning phase. Here, a process during operation start, a process during operation, and a process during operation stop performed by the learning unit 140a will be described.

  For example, the learning unit 140a acquires the vibration information from the sensor 50, specifies the time when the strength of the vibration information is equal to or greater than the threshold value as the start time, and operates the time from the start time to a predetermined time later. It is determined that it is starting. The learning unit 140a determines that the preset time from the first time to the second time is in operation. The learning unit 140a stops the operation from the third time to the fourth time when the third time is set as the third time before the fourth time when all of the components 10a to 10e of the machine 40 are stopped. Judged as medium.

  The process during the operation start which the learning part 140a performs is demonstrated. The learning unit 140a acquires first vibration information during operation start from the normal vibration table 130a, and acquires vibration information during operation start from the sensor 50. The learning unit 140a compares the first vibration information with the vibration information acquired from the sensor 50, and if the vibration information does not match, the learning unit 140a displays on the display unit 120 that an error has occurred during the start of operation. Let

  The learning unit 140a displays the fact that an error has occurred during the start of operation on the display unit 120, and then receives input of cause information from the input unit 110. The learning unit 140a registers the condition “being started”, the second vibration information, and the cause information in association with each other in the learning table 130b.

  FIG. 8 is a diagram for explaining processing during the start of operation of the learning unit. It is assumed that the learning unit 140a has acquired the vibration information 20 as illustrated in FIG. When the learning unit 140a compares the first vibration information illustrated in FIG. 4 and the vibration information 20, the vibration information does not match in the region 20a. In this case, the learning unit 140a causes the display unit 120 to display that an error has occurred during the start of operation. Then, the learning unit 140a associates the condition “being started”, the vibration information 20, and the cause information received from the input unit 110, and registers them in the learning table 130b.

  Processing during operation performed by the learning unit 140a will be described. The learning unit 140a acquires first operating vibration information from the normal vibration table 130a, and acquires operating vibration information from the sensor 50. The learning unit 140a compares the first vibration information with the vibration information acquired from the sensor 50, and if the vibration information does not match, the learning unit 140a displays on the display unit 120 that an error has occurred during operation. .

  The learning unit 140a displays on the display unit 120 that an error has occurred during operation, and then receives input of cause information from the input unit 110. The learning unit 140a registers the condition “operating”, the second vibration information, and the cause information in association with each other in the learning table 130b.

  FIG. 9 is a diagram for explaining processing during operation of the learning unit. The learning unit 140a acquires vibration information 20 as illustrated in FIG. 9 from the sensor 50 during operation. When the learning unit 140a compares the first vibration information illustrated in FIG. 5 with the vibration information 21, the pieces of vibration information do not match in the region 21a. In this case, the learning unit 140a causes the display unit 120 to display that an error has occurred during operation. Then, the learning unit 140a associates the condition “in operation”, the vibration information 21, and the cause information received from the input unit 110 and registers them in the learning table 130b.

  The process during operation stop performed by the learning unit 140a will be described. The learning unit 140a acquires the first vibration information during operation stop from the normal vibration table 130a, and acquires the vibration information during operation stop from the sensor 50. The learning unit 140a compares the first vibration information with the vibration information acquired from the sensor 50, and if the vibration information does not match, the learning unit 140a displays on the display unit 120 that an error has occurred during operation stop. Let

  The learning unit 140a displays the fact that an error has occurred during operation stop on the display unit 120, and then receives input of cause information from the input unit 110. The learning unit 140a registers the condition “not working”, the second vibration information, and the cause information in association with each other in the learning table 130b.

  FIG. 10 is a diagram for explaining a process during the operation stop of the learning unit. It is assumed that the learning unit 140a acquires the vibration information 20 as illustrated in FIG. When the learning unit 140a compares the first vibration information illustrated in FIG. 6 with the vibration information 22, the vibration information does not match in the region 22a. In this case, the learning unit 140a causes the display unit 120 to display that an error has occurred during the operation stop. Then, the learning unit 140a associates the condition “in operation stop”, the vibration information 22, and the cause information received from the input unit 110, and registers them in the learning table 130b.

  Note that the learning unit 140a described above may determine whether or not the vibration information acquired from the sensor 50 matches the first vibration information by calculating the similarity of each vibration information. For example, the learning unit 140a determines that the vibration information matches when the similarity of the vibration information is greater than a predetermined similarity.

  Returning to the description of FIG. The identification unit 140b is a processing unit that identifies the cause of the abnormality when it is determined that the vibration information of the sensor 50 is abnormal in the specific phase. Here, a process during operation start, a process during operation, and a process during operation stop performed by the specifying unit 140b will be described. Note that the process of determining whether the specifying unit 140b is in operation, in operation, or in operation is the same as the above-described process in which the learning unit 140a determines whether the operation is in operation, in operation, or in operation is stopped.

  A process during operation start executed by the specifying unit 140b will be described. The specifying unit 140b acquires first vibration information during operation start from the normal vibration table 130a, and acquires vibration information during operation start from the sensor 50. The specifying unit 140b compares the first vibration information with the vibration information acquired from the sensor 50, and determines that the vibration information of the sensor 50 is abnormal when the pieces of vibration information do not match. The vibration information determined by the specifying unit 140b to be abnormal is referred to as third vibration information.

  The specifying unit 140b compares the second vibration information associated with the condition “being started” in the learning table 130b with the third vibration information, and specifies the second vibration information that matches the third vibration information. . The identifying unit 140b may determine that the second vibration information that is most similar to the third vibration information among the second vibration information is the second vibration information that matches the third vibration information. The specifying unit 140b specifies cause information associated with the second vibration information that matches the third vibration information, and causes the display unit 120 to display the specified cause information.

  Processing during operation performed by the specifying unit 140b will be described. The specifying unit 140b acquires first operating vibration information from the normal vibration table 130a, and acquires operating vibration information from the sensor 50. The specifying unit 140b compares the first vibration information with the vibration information acquired from the sensor 50, and determines that the vibration information of the sensor 50 is abnormal when the pieces of vibration information do not match. The vibration information determined by the specifying unit 140b to be abnormal is referred to as third vibration information.

  The specifying unit 140b compares the second vibration information associated with the condition “in operation” in the learning table 130b with the third vibration information, and specifies the second vibration information that matches the third vibration information. The identifying unit 140b may determine that the second vibration information that is most similar to the third vibration information among the second vibration information is the second vibration information that matches the third vibration information. The specifying unit 140b specifies cause information associated with the second vibration information that matches the third vibration information, and causes the display unit 120 to display the specified cause information.

  The process during operation stoppage executed by the specifying unit 140b will be described. The specifying unit 140b acquires first vibration information during operation stop from the normal vibration table 130a, and acquires vibration information during operation stop from the sensor 50. The specifying unit 140b compares the first vibration information with the vibration information acquired from the sensor 50, and determines that the vibration information of the sensor 50 is abnormal when the pieces of vibration information do not match. The vibration information determined by the specifying unit 140b to be abnormal is referred to as third vibration information.

  The identifying unit 140b compares each second vibration information associated with the condition “operation stopped” in the learning table 130b with the third vibration information, and identifies the second vibration information that matches the third vibration information. . The identifying unit 140b may determine that the second vibration information that is most similar to the third vibration information among the second vibration information is the second vibration information that matches the third vibration information. The specifying unit 140b specifies cause information associated with the second vibration information that matches the third vibration information, and causes the display unit 120 to display the specified cause information.

  Next, a processing procedure of the specific apparatus 100 according to the present embodiment will be described. FIG. 11 is a flowchart illustrating the processing procedure of the specific device in the learning phase. As illustrated in FIG. 11, the learning unit 140a of the identification device 100 acquires vibration information from the sensor 50 (step S101). The learning unit 140a compares the vibration information with normal vibration information (first vibration information) (step S102).

  The learning unit 140a proceeds to step S107 when the vibration information acquired from the sensor 50 matches the vibration information at the normal time (step S103, Yes). On the other hand, when the vibration information acquired from the sensor 50 and the vibration information at the normal time do not match (No at Step S103), the learning unit 140a proceeds to Step S104.

  Step S104 will be described. The learning unit 140a causes the display unit 120 to output a warning (step S104). The learning unit 140a acquires cause information related to the part in which an abnormality has occurred (step S105). The learning unit 140a updates the learning table 130b (step S106), and proceeds to step S107.

  Step S107 will be described. The learning unit 140a determines whether or not to end the learning phase (step S107). The learning unit 140a ends the learning phase when the learning phase ends (Yes in step S107). If the learning unit 140a does not end the learning phase (No at Step S107), the learning unit 140a proceeds to Step S101.

  In the example illustrated in FIG. 11, the description has been given on the assumption that each time the warning is output by the specific device 100, the worker inputs the cause information regarding the part in which the abnormality has occurred. It is not limited. For example, when the worker inputs the cause information for the second vibration information collectively at a timing convenient for the worker such as the end of the day or the end of the week, the specific device 100 that has received the input receives the learning table 130b. May be updated.

  FIG. 12 is a flowchart showing the processing procedure of the specifying unit in the specific phase. As illustrated in FIG. 12, the specifying unit 140b of the specifying apparatus 100 acquires vibration information from the sensor 50 (step S201). The specifying unit 140b compares the vibration information with normal vibration information (first vibration information) (step S202).

  When the vibration information acquired from the sensor 50 matches the vibration information at the normal time (step S203, Yes), the specifying unit 140b proceeds to step S207. On the other hand, when the vibration information acquired from the sensor 50 and the vibration information at the normal time do not match (No at Step S203), the specifying unit 140b proceeds to Step S204.

  The identifying unit 140b compares each second vibration information in the learning table 130b with vibration information (third vibration information) acquired from the sensor 50 (step S204), and identifies cause information (step S205). The identification unit 140b outputs cause information (step S206).

  Step S207 will be described. The specifying unit 140b determines whether or not to continue the specific phase (step S207). The identification unit 140b ends the specific phase when the specific phase is not continued (No in Step S207). On the other hand, the identification part 140b transfers to step S201, when continuing a specific phase (step S207, Yes).

  Next, the effect of the specific device 100 according to the present embodiment will be described. In the learning phase, the identifying apparatus 100 learns the second vibration information and the cause information at the time of occurrence of an abnormality in the learning table 130b. And the specific apparatus 100 can notify cause information with the single sensor 50 by utilizing the learning table 130b in a specific phase. For this reason, since the number of sensors for specifying an abnormal part can be reduced, the cost of equipment for detecting an abnormality of the part can be reduced.

  The specific device 100 compares the vibration information acquired from the sensor 50 with the first vibration information during the operation start during the operation start, and the second vibration information during the operation start when the vibration information is different. The cause information is stored in the learning table 130b. For this reason, component information in which an abnormality has occurred during the start of operation can be specified by the single sensor 50.

  The specific device 100 compares the vibration information acquired from the sensor 50 during operation and the first vibration information during operation, and when each vibration information is different, the second vibration information during operation and the cause information Are stored in the learning table 130b. For this reason, component information in which an abnormality has occurred during operation can be specified by the single sensor 50.

  When the specific device 100 is operating, the vibration information acquired from the sensor 50 is compared with the first vibration information during operation stop, and when each vibration information is different, the second vibration information during operation stop and the cause Information is stored in the learning table 130b. For this reason, it is possible to identify the component information in which an abnormality has occurred during the operation stop by the single sensor 50.

  In the present embodiment, as an example, the specific apparatus 100 acquires vibration information using the single sensor 50, but the present invention is not limited to this. For example, the identification device 100 may install n sensors in the machine 40 and acquire vibration information from each sensor. n is a natural number of 2 or more and less than the number of types of parts included in the machine 40.

  Further, in this embodiment, as an example, the description has been made using the operation start, operation, and operation stop as conditions. However, by further using other conditions, each vibration information and cause information is subdivided to be normal. You may register into the vibration table 130a and the learning table 130b. For example, the type of material to be processed, the installation position of the sensor, the number of sensors, the day of the week when the machine 40 is operated, and the like may be further added to the conditions.

  Further, the specifying unit 140b of the specifying apparatus 100 may further specify whether or not there is a sign of a failure of the machine 40. For example, the specifying unit 140b compares the normal first vibration information with the vibration information measured by the sensor 50, and periodically determines whether or not the error ratio is equal to or greater than a threshold value. The identifying unit 140b counts the number of times that the error rate is equal to or greater than the threshold, and when the number is greater than or equal to the predetermined number, identifies that the machine 40 has a sign of failure and outputs a warning.

  Further, in the description of the present embodiment, it has been described that the sensor 50 measures vibration information and performs processing. However, for example, sound information is collected using a microphone or the like, and similar processing is performed. Also good.

  By the way, among the processes described in the present embodiment, all or a part of the processes described as being automatically performed can be manually performed, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedure, control procedure, specific name, and information including various data and parameters shown in the above-described document and drawings can be arbitrarily changed unless otherwise specified.

  Furthermore, each processing function performed by each device may be realized by a CPU and a program that is analyzed and executed by the CPU, or may be realized as hardware by wired logic.

DESCRIPTION OF SYMBOLS 50 Sensor 100 Identification apparatus 110 Input part 120 Display part 130 Storage part 130a Normal vibration table 130b Learning table 140 Control part 140a Learning part 140b Identification part

Claims (6)

Installed in a device having a plurality of individually operated components, and a measurement unit for measuring vibration information of the device;
The first vibration information measured by the measurement unit under the condition that all of the plurality of parts normally operate, and the measurement unit measured under the condition where it is unknown whether the plurality of parts are operating normally. When the first vibration information and the second vibration information are different from each other by comparing the second vibration information, the second vibration information and cause information indicating which component is the cause of the abnormality A learning unit that associates and registers in the table;
When the third vibration information measured by the measurement unit is different from the first vibration information, the cause information of the abnormality is identified based on the third vibration information and the second vibration information of the table. And a specific device.   When the learning unit outputs a warning when the first vibration information is different from the second vibration information and receives cause information of the abnormality from the input unit, the learning unit causes the second vibration information and the cause of the abnormality. The identification device according to claim 1, wherein a process of associating information with each other and registering the information in the table is repeatedly executed.   The learning unit compares the first vibration information and the second vibration information measured until a predetermined time after any one of the plurality of parts starts operation, and the first vibration information 3. The identification device according to claim 1, wherein when the second vibration information is different, the second vibration information and the cause information of the abnormality are registered in a table in association with each other.   The learning unit includes the first vibration information and the second vibration information measured from a time before a predetermined time from a time when all of the plurality of parts stop to a time when all of the plurality of parts stop. 2. If the first vibration information and the second vibration information are different from each other, the second vibration information and the cause information of the abnormality are registered in association with each other in the table. 2. The specific device according to 2 or 3.   The learning unit compares the first vibration information and the second vibration information measured in a time zone in which all operating states of the plurality of parts are in a steady state, and compares the first vibration information with the first vibration information. 5. The identification device according to claim 1, wherein when the second vibration information is different, the second vibration information and the cause information of the abnormality are registered in a table in association with each other. A specific method performed by a computer,
Using a measuring device installed in a device having a plurality of individually operated parts, vibration information of the device is measured, and the first measured by the measuring device under the condition that all the plurality of parts operate normally. The vibration information is compared with the second vibration information measured by the measuring device under a condition where it is unknown whether the plurality of parts are operating normally. The first vibration information and the second vibration information Are registered in the table in association with the second vibration information and cause information indicating which component is the cause of the abnormality,
When the third vibration information measured by the measurement device is different from the first vibration information, the cause information is specified based on the third vibration information and the second vibration information of the table. The specific method characterized by performing.

Images