JP5081999B1 - How to display abnormal sign diagnosis results - Google Patents

Abstract

An abnormality sign diagnosis result display method for appropriately displaying information including a diagnosis result by an abnormality sign diagnosis device for diagnosing presence / absence of an abnormality sign in a mechanical facility is provided.
A first diagnostic procedure for learning a normal model indicating a normal range of sensor data and diagnosing presence / absence of an abnormality sign of a mechanical facility based on the normal model, and a predetermined range in which sensor data values are set in advance A second diagnostic procedure for diagnosing that there is a sign of abnormality when the threshold value of the threshold is exceeded, and a display control procedure for controlling the display of information including the diagnostic result. It is possible to identify the period during which the maintenance is performed, the predetermined learning period after the maintenance work, the diagnosis period after the predetermined learning period until the next maintenance work is performed, and the presence or absence of a sign of abnormality in the mechanical equipment during the diagnosis period Is displayed on the display means.
[Selection] Figure 17

Description

  The present invention relates to a method for displaying an abnormality sign diagnosis result by an abnormality sign diagnosis device that diagnoses the presence or absence of an abnormality sign based on sensor data from mechanical equipment.

Mechanical equipment and system equipment composed of various devices are required to operate stably and normally. However, various troubles occur in mechanical equipment and system devices due to load fluctuations based on changes in operating conditions and other factors.
Normally, these devices are provided with various sensors for the purpose of grasping the operation state and detecting an abnormality, and monitoring the operation state. Furthermore, recent advances in IT equipment enable the storage of enormous operation record data and sensor data, and these data are stored in many devices and systems.
In addition, for example, in a power generation system or the like, it is required to operate for 24 hours without stopping, and for that purpose, periodic inspection and maintenance for preventing the occurrence of a failure are indispensable. Furthermore, if a sign before the occurrence of an abnormal state can be detected, an early response can be made by analyzing sensor data before and after the sign.

Conventionally, two methods have been proposed as methods for detecting an anomaly sign. The first method is predictive abnormality diagnosis by remote monitoring with threshold setting. That is, a threshold value is set in advance for each sensor data, and when the sensor data value exceeds a predetermined threshold value, it is determined that there is a sign of abnormality.
Patent Document 1 includes measurement means for measuring indoor temperature, gas concentration, radiation, and the like, and determination means for determining the state of the equipment in the room, and the determination means has a level of a measurement value from each measurement means. It describes a technique for discriminating three states of indoor normality, indoor attention state, and indoor abnormality from the size and the combination thereof. According to the technique described in Patent Literature 1, an abnormal sign such as mechanical equipment can be detected by appropriately setting a threshold value of the measurement value.

  The second method for detecting an abnormal sign is an abnormal sign diagnosis by data mining using a statistical classification method. Patent Document 2 describes a technique for setting a normal range as a clustering map and detecting an abnormal sign using a competitive neural network. According to the technique described in Patent Document 2, it has excellent detection accuracy even when there is a variation in the normal state, and can detect the abnormal state of the monitoring target without requiring data at the time of abnormality.

JP 11-125694 A JP 2007-198918 A

In the technique described in Patent Literature 1, when detecting an abnormal sign, there are a small number (for example, one type) of sensors exceeding the threshold value. However, there are many complex causes of abnormalities that occur in complex machinery and equipment. For example, even if the failure items are the same, there are various processes leading to the failure, and it is not always possible to set an optimal threshold value. There is a problem.
Further, in the technique described in Patent Document 2, for example, when maintenance is performed on mechanical equipment or the like, the system conditions may change greatly. Even in such a case, if a diagnosis based on data mining is performed using the accumulated past data as it is, there is a problem that a diagnosis is made as a sign of abnormality even though the mechanical equipment itself is normal.
Further, for example, a technique for performing learning by data mining based on maintenance information input from the outside and automatically displaying a diagnosis result based on the learning and a schedule on which maintenance is performed is disclosed in Patent Literature It is not described in 2.

  Then, this invention makes it a subject to provide the display method of the abnormal sign diagnostic result which displays appropriately the information containing the diagnostic result by the abnormal sign diagnostic apparatus which diagnoses the presence or absence of the abnormal sign of a mechanical installation.

  In order to solve the above-mentioned problems, the present invention provides a sensor data storage means for storing sensor data from various sensors acquired at predetermined sampling periods, and a maintenance information storage means for storing maintenance information of mechanical equipment. A maintenance information acquisition procedure for acquiring maintenance information from the maintenance information storage means, including at least information for identifying the mechanical equipment and maintenance work period in which the maintenance work is performed, and maintenance information. Corresponding to the maintenance information acquired by the acquisition procedure, the learning data acquisition procedure for specifying the period during which the sensor data was acquired and reading the sensor data corresponding to the period from the sensor data storage means, and the normality of the sensor data A learning procedure for learning a normal model indicating a range, a diagnostic procedure for diagnosing the presence or absence of an abnormality sign of mechanical equipment based on the normal model, A first diagnostic step comprising: reading out the sensor data from the sensor data storage means; and if each sensor data exceeds a predetermined threshold value set in advance, a second diagnosis is made that there is a sign of abnormality And a display control process for performing display control of information including the diagnosis result of the first diagnosis process and the diagnosis result of the second diagnosis process. In the display control process, at least maintenance work is performed. It is possible to identify the period during which the maintenance is performed, the predetermined learning period after the maintenance work, the diagnosis period after the predetermined learning period until the next maintenance work is performed, and the presence or absence of a sign of abnormality in the mechanical equipment during the diagnosis period And displaying on the display means.

  According to the present invention, it is possible to provide a method for displaying an abnormality sign diagnosis result that appropriately displays information including the diagnosis result by using an abnormality sign diagnosis device that diagnoses the presence or absence of an abnormality sign of mechanical equipment.

1 is a system configuration diagram including an abnormality sign diagnosis apparatus according to an embodiment of the present invention. It is a figure which shows the example of the registration screen of the maintenance information displayed on the display means of a computer. It is a block diagram of the data mining part of the abnormality sign diagnostic apparatus. It is explanatory drawing which shows the relationship between the learning period containing the learning object data acquired by the learning object data acquisition means, and a diagnosis date. It is explanatory drawing which shows the relationship between the learning period containing the learning object data acquired by the learning object data acquisition means, and a diagnosis date, when a maintenance work is performed. It is a figure which shows the change of the value of the various sensor data in each operation mode. It is explanatory drawing of learning of the normal model by a data mining learning part. It is a figure which shows the example of the learning result by a data mining learning part. It is a flowchart which shows the learning process in a data mining learning part. It is a flowchart which shows the acquisition process of the learning object data by a learning object data acquisition means. It is a flowchart which shows the diagnostic process in a data mining diagnostic part. It is a figure which shows the example of the diagnostic result by a data mining diagnostic part. It is a block diagram of the remote monitoring part of the abnormality sign diagnostic apparatus. It is explanatory drawing of the diagnosis of the abnormal sign by the separate determination means of a remote monitoring part. It is a figure which shows the example of the experimental data by an abnormality sign diagnostic apparatus, (a) shows the change of a measured value, a learning value, and a contribution, (b) is a graph which shows the change of an abnormal degree. (A) is a graph which shows the change of the measured value in the period before and behind performing maintenance work, and the change of the abnormal value in a comparative example, (b) is the case where the abnormality sign diagnostic apparatus which concerns on this embodiment is used It is a graph which shows the change of the abnormal value in the period before and after performing maintenance work. It is a figure which shows the example at the time of displaying the diagnostic result by the abnormality sign diagnostic apparatus on a display means. It is a figure which shows the example at the time of displaying the diagnostic result by a remote monitoring part on a display means, (a) shows the abnormal sign detection screen at the time of detecting an abnormal sign, (b) displays the content of the abnormal sign It is a figure which shows the example of the abnormal sign detail screen to do. It is a figure which shows the example at the time of displaying the change of the measured value by the some sensor installed in the mechanical installation on a display means. It is a flowchart which shows the display process by a display control means.

  Embodiments of the present invention will be described in detail with reference to the drawings as appropriate. In each figure, common portions are denoted by the same reference numerals, and redundant description is omitted.

<< First Embodiment >>
FIG. 1 is a system configuration diagram including an abnormality sign diagnosis apparatus according to an embodiment of the present invention.
The abnormality sign diagnosis device 1 is a device that diagnoses whether there is an abnormality sign in the machine facility 2 based on measurement values acquired from a plurality of sensors installed in the machine facility 2. Here, the “abnormal sign” indicates that there is a high possibility that an abnormal situation will occur in the mechanical equipment 2 in the near future (for example, within approximately two weeks) with reference to the diagnosis date. Further, “abnormal” indicates, for example, a situation in which a malfunction (failure) of the operation is clearly generated in the mechanical equipment 2 and the mechanical equipment 2 is not operating normally.
Note that the boundary for distinguishing “abnormality sign” from “abnormality” is not necessarily clear. For example, a scale indicating the degree of abnormality (an abnormality degree described later) is appropriately set, and the value of the scale is The two may be distinguished depending on whether or not a predetermined threshold value is exceeded.

As shown in FIG. 1, the abnormality sign diagnosis apparatus 1 according to the present embodiment is connected to at least one mechanical facility 2 via a communication network N. Usually, there are a plurality of machine facilities 2 (for example, 50), but even in such a case, in the following description, “machine facilities 2” will be collectively described. Incidentally, the mechanical equipment 2 indicates, for example, gas engines and gas turbines used in power plants, nuclear reactors in nuclear power plants, windmills in wind power plants, and various devices used in chemical plants.
In addition, in order to measure temperature, pressure, voltage, current, motor rotation speed, etc., a plurality of (not shown) sensors (for example, 30 sensors) are installed in the mechanical equipment 2, and a certain sample period ( For example, each measurement value is acquired every 30 seconds. Incidentally, in the transient state of the mechanical equipment 2, it is preferable to set the sample period shorter.

The machine facility 2 transmits the measured values acquired by the various sensors described above to the abnormality sign diagnosis apparatus 1 via the communication network N.
Further, each machine facility 2 is provided with a plurality of interlocks (not shown) as safety instrumentation in each machine facility 2 itself, in addition to being subjected to an abnormality sign diagnosis by the abnormality sign diagnostic apparatus 1. . In a predetermined case (for example, when the amount of fuel is low), alarm information or failure information from the interlock is transmitted to the abnormality sign diagnosis apparatus 1 via the communication network N. Yes.

In addition, maintenance work may be performed on the mechanical equipment 2. The maintenance work may be performed as regular maintenance, or alarm information or failure information from an interlock (not shown) installed in the mechanical equipment 2 itself, or diagnosis of an abnormal sign by the abnormal sign diagnosis device 1 It may be performed as maintenance corresponding to Further, the maintenance work includes, for example, an operation for adding cooling water from the outside when the mechanical equipment 2 is a gas engine.
In other words, “maintenance work” means an artificial operation including maintenance on the mechanical equipment 2. Below, the case where the maintenance with respect to the mechanical installation 2 is performed is demonstrated as an example of a maintenance operation | work.

In addition, as shown in FIG. 1, the abnormality sign diagnosis apparatus 1 is connected to a computer 3 via a communication network N. The computer 3 includes input means, control means, storage means, display means, communication means, etc. (not shown). Then, the worker who has grasped the content of the maintenance (maintenance information) for each mechanical facility 2 can use the input means (keyboard or the like) of the computer 3 to perform identification information and maintenance work period of the mechanical facility 2 in which the maintenance operation is performed. Maintenance information including the above is registered (see FIG. 2).
FIG. 2 is a diagram showing an example of a maintenance information registration screen displayed on the display means (not shown) of the computer.
The maintenance information input to the computer 3 by the administrator when performing maintenance on the mechanical equipment 2 includes, for example, the following. That is, as shown in FIG. 2, as registration basic information 301, a subject name, a correspondence type, a site name that manages the machine facility 2, and the like are input. Incidentally, “management No.”, “site name”, and “unit” in the basic registration information 301 shown in FIG. 2 identify the machine facility 2 and the site that manages the machine facility 2.

In addition, as the defect information 302, the date and time when a defect such as a failure occurs in the mechanical equipment 2, the state, the error item (failure occurrence location), and the like are input.
In addition, as the maintenance work period 303, a period during which maintenance is to be performed on the machine facility 2 in which a failure has occurred is input. That is, the maintenance work period 303 input to the computer 3 is information indicating a period during which maintenance is scheduled to be performed on the mechanical equipment 2 in the future (the current time when the maintenance information is input is assumed to be present).
Incidentally, the maintenance work period 303 input to the computer 3 may be information indicating that the maintenance has been performed on the machine facility 2 in a predetermined period in the past (the time when the maintenance information is input is the present). In the latter case, the abnormality sign diagnosis apparatus 1 again executes the diagnosis of the abnormality sign based on the information that the maintenance has been performed in the past predetermined period.

As the replacement part 304, information for specifying the part of the mechanical equipment 2 replaced at the time of maintenance is input. Further, as information 305 indicating whether or not the maintenance work is completed, information indicating whether or not the maintenance work is completed is input corresponding to 0 or 1. Further, as the treatment content 306, the state of the mechanical equipment 2 before and after the maintenance is input in addition to the maintenance content.
Maintenance information input to the computer 3 is stored in storage means (not shown) of the computer 3 by control means (not shown), and further transmitted to the abnormality sign diagnosis apparatus 1 via the communication network N.

<Configuration of abnormal sign diagnosis device>
As shown in FIG. 1, the abnormality sign diagnostic apparatus 1 includes a communication unit 10, a sensor data acquisition unit 11, a sensor data storage unit 12, a maintenance information acquisition unit 13, a maintenance information storage unit 14, and a data mining unit. 15, a remote monitoring unit 16, a diagnosis result storage unit 17, a display control unit 18, and a display unit 19.
The communication means 10 receives data from the mechanical equipment 2 and the computer 3 via the communication network N, and is constituted by, for example, a router or various interfaces. The communication means 10 receives the data via the communication network N, for example, according to a TCP / IP communication protocol.

  The sensor data acquisition unit 11 acquires sensor data from various sensors (not shown) installed in the machine facility 2 among the data input from the communication unit 10 via the communication network N. As described above, the various sensors installed in the mechanical equipment 2 measure predetermined output values (for example, temperature and pressure values) in the mechanical equipment 2. The mechanical equipment 2 acquires the measurement values obtained by the various sensors every predetermined time (for example, 30 seconds), and sequentially transmits them to the abnormality sign diagnosis apparatus 1 via the communication network N. In this case, the sensor data acquisition unit 11 acquires 2880 measurement values from the various sensors of the machine facility 2 per day.

In addition to the measurement value, the data input from the communication means 10 via the communication network N includes the data type of the measurement value detected by the sensor (the identification number of the mechanical equipment 2 and the sensor identification). Including the date) and the date and time when the measured value was acquired in the mechanical equipment 2. In addition, as data input from the communication unit 10, a predetermined command value (for example, a motor rotation speed command value) in the mechanical equipment 2 may be included.
Hereinafter, the measurement value and the data added to specify the measurement value are collectively referred to as “sensor data”. The sensor data acquisition unit 11 stores the sensor data acquired through the communication unit 10 in the sensor data storage unit 12.

  The sensor data storage unit 12 stores the sensor data acquired by the sensor data acquisition unit 11. That is, the sensor data storage unit 12 stores sensor data as a database for each site (not shown) that manages the machine facility 2 or at least one machine facility 2. The sensor data storage unit 12 is configured to be able to output the stored sensor data to a data mining unit 15, a remote monitoring unit 16, and a display control unit 18, which will be described later.

  The maintenance information acquisition unit 13 acquires maintenance information output from the computer 3 out of each data input from the communication unit 10 via the communication network N. The maintenance information acquired by the maintenance information acquisition unit 13 includes at least an identification number (see registration basic information 301 in FIG. 2) for specifying the machine equipment 2 and maintenance work performed on the machine equipment 2. Period (see maintenance work period 303 in FIG. 2). The maintenance information acquisition unit 13 identifies whether the data acquired through the communication unit 10 is maintenance information, and stores the acquired maintenance information in the maintenance information storage unit 14.

The maintenance information storage unit 14 stores the maintenance information acquired by the maintenance information acquisition unit 13. The maintenance information storage unit 14 stores, as a database, maintenance information including at least for which machine equipment 2 maintenance work has been performed for which period. The maintenance information storage unit 14 is configured to be able to output the stored maintenance information to the data mining unit 15 that will be described later and the display control unit 18.
The data mining unit 15 performs data mining by applying a statistical data classification method, learns normal data of the machine equipment 2 and learns a normal model for each machine equipment 2, and based on the normal model, the machine equipment 2 Diagnose the presence or absence of abnormal signs. Further, the data mining unit 15 acquires sensor data from the sensor data storage unit 12 based on the maintenance information acquired from the maintenance information storage unit 14. The data mining unit 15 performs learning and diagnosis based on the acquired sensor data, and stores the diagnosis result in the diagnosis result storage unit 17.
Details of the data mining unit 15 will be described later.

The remote monitoring unit 16 sets an upper and lower threshold for the value (or amount of change) of the sensor data acquired from the mechanical equipment 2 based on the experience and knowledge of the administrator who performs various settings of the abnormality sign diagnosis device 1. When the sensor data value exceeds the upper and lower thresholds, it is diagnosed that there is a sign of abnormality. The remote monitoring unit 16 stores the diagnosis result in the diagnosis result storage unit 17.
Details of the remote monitoring unit 16 will be described later.
The abnormality sign diagnosis apparatus 1 independently performs diagnosis by the data mining unit 15 and diagnosis by the remote monitoring unit 16.

The diagnosis result storage means 17 stores the diagnosis result by the data mining unit 15 and the diagnosis result by the remote monitoring unit 16 in separate storage areas (not shown). The diagnosis result storage unit 17 stores each diagnosis result on the diagnosis date for each machine facility 2 (or a site that manages the machine facility 2), and is configured to be output to the display control means 18. .
The display control unit 18 performs control for causing the display unit 19 to display the diagnosis result by the data mining unit 15 and the diagnosis result by the remote monitoring unit 16 stored in the diagnosis result storage unit 17. That is, the display control means 18 outputs a control signal required when displaying the diagnosis result by the abnormality sign diagnosis apparatus 1 on the display means 19 (details will be described later).
The display unit 19 is, for example, a monitor, and displays a diagnosis result by the abnormality sign diagnosis device 1 according to a control signal input from the display control unit 18.

  FIG. 3 is a configuration diagram of a data mining unit of the abnormality sign diagnosis apparatus. The data mining unit 15 includes a data mining learning unit 151 and a data mining diagnostic unit 152.

<Configuration of data mining learning unit>
The data mining learning unit 151 performs data mining by applying clustering, which is a kind of statistical data classification method, and learns a normal model of the mechanical equipment 2. The target indicated by the “normal model” will be described later.
The data mining learning unit 151 includes a learning target data acquisition unit 151a, a mode determination unit 151b, a mode-specific learning target data storage unit 151c, a learning unit 151d, and a normal model data storage unit 151e.

(1. Learning object data acquisition means)
As shown in FIG. 3, the learning target data acquisition unit 151 a acquires sensor data from the sensor data storage unit 12 based on the maintenance information stored in the maintenance information storage unit 14. Hereinafter, the sensor data that the learning target data acquisition unit 151a acquires from the sensor data storage unit 12 is referred to as “learning target data”.

(A. Learning target data acquisition process in a normal case)
FIG. 4 is an explanatory diagram showing the relationship between the learning period including the learning target data acquired by the learning target data acquisition means and the diagnosis date. Incidentally, FIG. 4 shows a case where the maintenance of the mechanical equipment 2 is not performed from the first day (1 day) of the learning period B to the diagnosis date (22 days). Further, the date shown in FIG. 4 is given with the first day of the learning period B as a reference (one day).
As shown in FIG. 4, the learning target data acquisition unit 151a goes back to the past for a predetermined period A (for example, one week) from the diagnosis date (22 days), and learns for a predetermined learning period B (for example, two weeks). Target data is acquired from the sensor data storage means 12. Here, the predetermined period A and the learning period B are preset periods.
In addition, the predetermined period A can correspond to the detection of an event in which the equipment state changes abnormally slowly over a long period of time, and also can correspond to the change in the operating status of the mechanical equipment 2 itself. What is necessary is just to set suitably. The learning period B may be a period during which the learning means 151d described later can sufficiently acquire learning target data necessary for learning a normal model, and may be set as appropriate.

Further, the learning target data acquisition unit 151a is set to exclude the day when an alarm or failure is detected by an interlock (not shown) installed in the machine facility 2 as the predetermined period B. That is, since there was an alarm in the mechanical facility 2 on the 2nd shown in FIG. 4, the learning target data acquisition unit 151a acquires sensor data for two weeks excluding 2 days from the sensor data storage unit 12 as learning target data. .
The learning process by the data mining learning unit 151 is performed once a day at a preset time (for example, midnight). A diagnosis process by the data mining diagnosis unit 152 described later is diagnosed once a day with reference to a normal model learned by the learning process after the learning process is completed.
The predetermined period A and the learning period B shown in FIG. 4 are shifted one day at a time as the diagnosis date elapses.
The learning process and the diagnosis process described above are not limited to once a day, and may be performed every predetermined period. In addition, the administrator of the abnormality sign diagnosis apparatus can arbitrarily set the timing for performing the learning process and the diagnosis process via an input unit (not shown).

Incidentally, in the above description, the day on which an alarm or failure is detected by an interlock (not shown) installed in the mechanical equipment 2 itself is excluded from the learning target day. It is preferable to exclude the day diagnosed as having a sign from the learning target. This is because the data mining unit 15 needs to learn normal data in order to perform highly accurate diagnosis using clustering. In this way, the sensor data of the mechanical equipment 2 on the day when the alarm or failure is detected and on the day excluding the day diagnosed as the abnormal sign is referred to as “normal data”.
That is, the learning target data acquisition unit 151a acquires normal data for a predetermined number of days in the past from the sensor data storage unit 12, and outputs the normal data to the mode determination unit 151b.

(B. Learning target data acquisition process when there is maintenance)
FIG. 5 is an explanatory diagram showing a relationship between a learning period including learning target data acquired by the learning target data acquisition unit and a diagnosis date when maintenance work is performed.
As shown in FIG. 5, for example, it is assumed that maintenance is performed on a specific machine facility 2 during a period of 3 to 5 days. In addition, the maintenance period (3 to 5 days) illustrated in FIG. 5 includes a period until the operation of the mechanical equipment 2 is stabilized after the maintenance work is completed.

When such maintenance is performed, there is a high possibility that the sensor data value will fluctuate significantly before and after the maintenance. If the learning target data acquisition unit 151a acquires the learning target data by the method shown in FIG. 4 applied in a normal case when the diagnosis date is immediately after the maintenance, the following situation occurs. To do. That is, the learning target data acquisition unit 151a acquires sensor data before maintenance and sensor data under maintenance as learning target data. Therefore, in this case, there is a possibility that the abnormal sign diagnostic apparatus 1 may continue to diagnose that there is an abnormal sign, even though the mechanical equipment 2 itself that has undergone maintenance is normal (see FIG. 16A). .
Therefore, when there is maintenance, the learning target data acquisition unit 151a acquires the sensor data of the learning period shown in FIG. 5 as learning target data.

As illustrated in FIG. 5, the learning target data acquisition unit 151a obtains sensor data before maintenance (before two days) and sensor data for three to five days during the maintenance period from learning target data to be acquired. exclude. That is, the learning target data acquisition unit 151a acquires sensor data of a predetermined learning period B1 (for example, one week) as learning target data from 6 days after the maintenance period (3 days to 5 days) has elapsed.
Note that the learning period B1 shown in FIG. 5 is a minimum period required for the learning unit 151d to create a new normal model using the learning target data acquired from the machine equipment 2 after the maintenance. Therefore, as shown in FIG. 5, the diagnostic means 152e (see FIG. 3) to be described later is an abnormal sign during the maintenance period (3 to 5 days) and the learning period B1 after maintenance (6 to 12 days). Do not diagnose. Then, the diagnosis unit 152e resumes the diagnosis of the abnormal sign from the 13th day after the learning period B1 has elapsed.

Then, as shown in FIG. 5A, until the learning period B2 (for example, two weeks) that is a sufficient period as the learning period can be secured, the learning target data acquisition unit 151a starts from the sixth day after the maintenance to the diagnosis date. Is acquired from the sensor data storage unit 12 as learning target data.
Further, as shown in FIG. 5B, until the number of days from the last day (19th day) of the learning period B2 to the diagnosis date reaches a predetermined period A2 (for example, one week), the learning object data acquisition unit 151a Then, sensor data in the learning period B2 is acquired. This is because, as described above, it is possible to cope with detection of an event in which the equipment state changes abnormally slowly over a long period of time.
Then, after the 27th day shown in FIG. 5, the learning target data acquisition unit 151a performs the same processing as the “learning data acquisition processing in the normal case” described above (see FIGS. 4 and 5B). . Further, as shown in FIG. 5 (c), from the diagnosis date (28th) when the number of days from the last day (19th) of the learning period B2 exceeds the predetermined period A2, the learning object data acquiring unit 151a performs the diagnosis. The learning period B3 and the predetermined period A3 are sequentially shifted with the passage of days.

By the way, a day when an alarm or a failure is detected by an interlock (not shown) installed in the mechanical equipment 2 itself or a day diagnosed by the abnormality sign diagnosis apparatus 1 as having an abnormality sign is excluded from the learning target days. This is the same as in the case of the “learning target data acquisition process in the normal case” described above.
Further, the settings of the learning periods B1 and B2 and the predetermined period A2 can be changed as appropriate.

(2. Mode determination means and mode-specific learning target data storage means)
The mode determination unit 151b determines the operation mode of the learning target data input from the learning target data acquisition unit 151a. Here, the “operation mode” means an operation mode of the mechanical equipment 2 (see FIG. 1).
FIG. 6 is a diagram illustrating changes in values of various sensor data in each operation mode.
As shown in FIG. 6, the operation mode is a mode in which the mechanical equipment 2 is stopped and in a steady state (steady state (OFF)), and a mode in which the mechanical equipment 2 is activated and in a transient state (transient ( START)), a mode in which the mechanical equipment 2 is operating and in a steady state (steady state (ON)), a mode in which the mechanical equipment 2 is stopped and in a transient state (transient (STOP)), and the like. The mode determination unit 151b can automatically determine the operation mode of the machine facility based on the change of the sensor data, and can divide the operation mode into each operation mode. For example, the mode determination unit 151b determines the operation mode of the mechanical equipment 2 according to the rate of change of the sensor data on the time axis, focusing on specific sensor data.

The mode determination unit 151b determines which operation mode the learning target data acquired by the learning target data acquisition unit 151a belongs to. And the mode determination means 151b matches the said operation mode with the learning object data acquired by the learning object data acquisition means 151a, and memorize | stores it in the learning object data storage means 151c (refer FIG. 3) according to mode.
The mode-specific learning target data storage unit 151c stores the learning target data input from the mode determination unit 151b for each operation mode. And the learning object data storage means 151c classified by mode is comprised so that the learning object data memorize | stored according to the operation mode can be output with respect to the learning means 151d.

(3. Learning means)
The learning unit 151d learns the normal model of the machine facility 2 using the learning target data acquired from the mode-specific learning target data storage unit 151c. The learning by the learning unit 151d is performed by performing data mining by applying clustering, which is a kind of statistical data classification method.

FIG. 7 is an explanatory diagram of normal model learning by the data mining learning unit. The learning unit 151d acquires learning target data of a specific operation mode determined as a learning target from the learning target data as normal data described above from the mode-specific learning target data storage unit 151c. The learning unit 151d classifies the learning target data into several representative groups called clusters for each similar data. In the example shown in FIG. 7, the case of one cluster is shown. However, a plurality of clusters may exist for one operation mode of a specific machine facility 2.
Further, the features α, feature β, and feature γ shown in FIG. 7 are normalized, for example, feature quantities (temperature, pressure, rotational speed, etc.) measured by a sensor (not shown) installed in the mechanical equipment 2. It is a thing. FIG. 7 shows a three-dimensional case as an example, but it is assumed that there are as many dimensions as the number of sensors acquired by the sensor data acquisition unit 11 (see FIG. 1).

As a clustering method, for example, a non-hierarchical clustering k-means method can be used. In the k-average method, first, clusters are randomly assigned to each data, and the center of each cluster (cluster center) is calculated based on the assigned data. As the cluster center, for example, the average of each element of the allocated data can be used. Next, the distance between the predetermined data and the cluster center is obtained, and the data is assigned to the nearest center cluster. If the cluster assignment of each data does not change for all data in such a process, the process ends. Otherwise, the cluster center is recalculated from the newly allocated cluster and the above process is repeated. Through the above processing, the learning unit 151d can classify the learning target data into a plurality of clusters.
Then, as shown in FIG. 7, a cluster center c and a cluster radius r are determined for each cluster. The cluster center c and the cluster radius r obtained for each cluster are normal models learned by the learning unit 151d. The learning unit 151d stores the learned normal model in the normal model data storage unit 151e (see FIG. 3) for each operation mode described above. Incidentally, in the abnormality sign diagnosis apparatus 1 according to the present embodiment, the learning time required for the learning means 151d per machine facility 2 is about 3 seconds.
Note that the learning unit 151d is set to learn a normal model when there is a predetermined number or more of learning target data acquired for each operation mode via the mode determination unit 151b. That is, the learning unit 151d does not learn the normal model when it is determined that the learning target data more than the minimum number necessary for learning the normal model does not exist in the mode-specific learning target data storage unit 151c. Information indicating that the diagnosis has not been performed is stored in the diagnosis result storage unit 17 (see FIG. 3).

(4. Normal model data storage means)
FIG. 8 is a diagram illustrating an example of a learning result by the data mining learning unit. As shown in FIG. 8, the normal model data storage unit 151e stores a normal model as a database for each operation mode. A cluster number 101 shown in FIG. 8 is an identification number assigned to identify each cluster. Sensors 1 to M (102) indicate sensors (not shown) installed in the machine facility 2.
The number of belonging data 103 shown in FIG. 8 indicates the number of learning target data belonging to each of the clusters 1 to N. The total error 104 indicates the total error (distance) between the cluster center of each cluster and each learning target data belonging to the cluster. The maximum error 105 is the maximum value of the distance between the cluster center of a specific cluster and a plurality of learning target data belonging to the cluster. The minimum error is a minimum value of the distance between the cluster center of a specific cluster and a plurality of learning target data belonging to the cluster.

The maximum value 107 indicates the maximum value of the feature component (for example, the coordinate value of the feature α shown in FIG. 7) corresponding to each sensor in the plurality of learning target data included in the specific cluster. The minimum value 108 indicates the minimum value of the feature component corresponding to each sensor in the plurality of learning target data included in the specific cluster. Incidentally, the learning unit 151d performs normalization processing using the values of the maximum value 107 and the minimum value 108. The cluster center 109 indicates the values of the sensors 1 to M that specify the center of each cluster.
In addition, the normal model memorize | stored in the normal model data storage means 151e is updated for every diagnosis day based on the learning object data which the learning object data acquisition means 151a acquires corresponding to a diagnosis date.

<Flow of learning processing by the data mining learning unit>
FIG. 9 is a flowchart showing a learning process in the data mining learning unit. In step S <b> 11, the learning target data acquisition unit 151 a (see FIG. 3) acquires learning target data from the sensor data storage unit 12.
FIG. 10 is a flowchart showing learning object data acquisition processing by the learning object data acquisition means. In step S21, the learning target data acquisition unit 151a determines whether or not the diagnosis date is within the normal period. Here, the “normal period” indicates a period when the learning period B and the predetermined period A shown in FIG. 4 are secured with reference to the date when the maintenance performed at the time closest to the diagnosis date is completed. .
When the diagnosis date is within the normal period (step S21 → Yes), the learning target data acquisition unit 151a acquires the learning target data from the sensor data storage unit 12 in the normal mode (step S22). Here, the “normal mode” indicates a mode in which the learning target data acquisition unit 151a acquires sensor data as learning target data in the learning period B that goes back in the past by a predetermined period A from the diagnosis date as shown in FIG. .

When the diagnosis date is not within the normal period (step S21 → No), the process of the learning target data acquisition unit 151a proceeds to step S23. In step S23, the learning target data acquisition unit 151a determines whether or not the diagnosis date is within the maintenance period. When the diagnosis date is within the maintenance period (step S23 → Yes), the process of the learning target data acquisition unit 151a proceeds to step S24. In step S24, the learning target data acquisition unit 151a stores information indicating that the diagnosis date is within the maintenance period in the diagnosis result storage unit 17 (see FIG. 3). Incidentally, in this case, since there is no sensor data that can be used as learning target data, the learning target data acquisition unit 151a does not acquire learning target data from the sensor data storage unit 12 (see FIG. 3).
On the other hand, when the diagnosis date is not within the maintenance period (step S23 → No), the processing of the learning target data acquisition unit 151a proceeds to step S25. In step S25, the learning target data acquisition unit 151a determines whether or not the diagnosis date is included in the learning period B1 (see FIG. 5). That is, the learning target data acquisition unit 151a determines whether or not a minimum learning period B1 is secured as a learning period after maintenance.

When the diagnosis date is included in the learning period B1 (step S25 → Yes), the processing of the learning target data acquisition unit 151a proceeds to step S26. In step S26, the learning target data acquisition unit 151a stores information indicating that the diagnosis date is within the learning period after maintenance in the diagnosis result storage unit 17 (see FIG. 3). In this case, because there is a shortage of sensor data that can be used as learning target data, the learning target data acquisition unit 151a does not acquire learning target data from the sensor data storage unit 12 (see FIG. 3).
On the other hand, when the diagnosis date is not included in the learning period B1 (step S25 → No), the processing of the learning target data acquisition unit 151a proceeds to step S27.

In step S27, the learning target data acquisition unit 151a determines whether the learning period B2 and the predetermined period A2 (see FIG. 5B) are secured. When the learning period B2 and the predetermined period A2 are secured (step S27 → Yes), the processing of the learning target data acquisition unit 151a proceeds to step S22. Incidentally, this case corresponds to the case of (b) or (c) in FIG. That is, in this case, it can be said that the diagnosis date corresponds to the “normal period” described above.
On the other hand, when the learning period B2 is not secured, or when the learning period B2 is secured but the predetermined period A2 is not secured (step S27 → No), the learning target data acquisition unit 151a is in the maintenance mode. Learning target data is acquired (step S28). Here, the “maintenance mode” is a transient learning period selection mode until returning to the “normal mode” as described with reference to FIG.

Returning to FIG. 9 again, in step S12, the mode determination unit 151b determines the operation mode of the learning target data. As described above, the mode determination unit 151b determines the operation mode of the mechanical equipment 2 based on the value or rate of change of specific learning target data.
In step S13, the mode determination unit 151b stores the learning target data for which the operation mode has been determined in the mode-specific learning target data storage unit 151c. Next, in step S14, the learning unit 151d selects an operation mode to be learned. For example, the learning unit 151d selects learning target data belonging to the steady (ON) mode (see FIG. 6) as the operation mode, and acquires it from the mode-specific learning target data storage unit 151c. Note that the operation mode selection order by the learning unit 151d is set in advance.

  In step S15 of FIG. 9, the learning unit 151d learns the normal model. In other words, as described above, the learning unit 151d classifies the learning target data into each cluster using a clustering method, and obtains the cluster center and the cluster radius. In step S16, the learning unit 151d stores the learning data corresponding to the normal model in the normal model data storage unit 151e. Next, in step S17, the learning unit 151d determines whether there is another operation mode to be learned. When there is another operation mode to be learned (step S17 → Yes), the process of the learning unit 151d returns to step S14. If there is no other operation mode to be learned (step S17 → No), the process of the data mining learning unit 151 returns.

<Configuration of data mining diagnosis unit>
The data mining diagnosis unit 152 shown in FIG. 3 diagnoses the presence or absence of an abnormality sign from the sensor data of the machine facility 2 based on the normal model generated by the data mining learning unit 151.
The data mining diagnosis unit 152 includes a diagnosis target data acquisition unit 152a, a mode determination unit 152b, a mode-specific diagnosis target data storage unit 152c, an abnormality degree calculation unit 152d, a diagnosis unit 152e, a contribution degree calculation unit 152f, Is provided.

  The diagnosis target data acquisition unit 152a acquires sensor data to be diagnosed (hereinafter referred to as diagnosis target data) from the sensor data storage unit 12. In the present embodiment, the diagnosis target data acquisition unit 152a uses the sensor data for the previous day as diagnosis target data once a day at a predetermined time (for example, midnight) as the diagnosis target data. 12 from. Incidentally, as described above, before the data mining diagnosis unit 152 performs the diagnosis process, the data mining learning unit 151 acquires the learning target data corresponding to the diagnosis date in advance (see FIGS. 4 and 5), and the learning data. Is stored in the normal model data storage means 151e.

The mode determination unit 152b determines the operation mode of the diagnosis target data input from the diagnosis target data acquisition unit 152a. The process of the mode determination unit 152b is the same as the process of the mode determination unit 151b described above with reference to FIG. The mode determination means 152b associates the operation mode with the diagnosis object data acquired by the diagnosis object data acquisition means 152a, and stores it in the mode-specific diagnosis object data storage means 152c.
The mode-specific diagnosis target data storage unit 152c stores the diagnosis target data input from the mode determination unit 152b for each operation mode. The mode-specific diagnosis target data storage unit 152b is configured to be able to output the diagnosis target data stored for each operation mode to the abnormality degree calculation unit 152d and the contribution degree calculation unit 152f.

The degree-of-abnormality calculation unit 152d reads out the diagnosis target data for each mode from the mode-specific diagnosis target data storage unit 152c, and calculates the degree of abnormality of the diagnosis target data. For example, it is assumed that the diagnosis target data exists at the position shown in FIG. 7 and the cluster center closest to the diagnosis target data is a cluster center c shown in FIG. In this case, the degree of abnormality u is expressed by the following equation (1) using the cluster radius r and the distance d from the cluster center c to the diagnosis target data.
u = d / r Formula (1)

The abnormality degree calculation means 152d outputs the abnormality degree u calculated using the equation (1) to the diagnosis means 152e. Further, the abnormality degree calculation unit 152d associates the diagnosis target data with the abnormality degree u and stores the data in the diagnosis result storage unit 17 (see FIG. 3). The diagnosis unit 152e diagnoses the presence or absence of an abnormality sign for the machine facility 2 (see FIG. 1) from which the diagnosis target data is acquired based on the abnormality degree u input from the abnormality degree calculation unit 152d.
When the degree of abnormality u ≦ 1, it can be said that the diagnosis target data exists in the cluster shown in FIG. That is, in this case, since the diagnosis target data exists within the range of the specific normal model, the diagnosis unit 152e determines that the corresponding machine facility 2 is normal (no abnormal sign).
On the other hand, when the degree of abnormality u> 1, it can be said that the diagnosis target data exists outside the cluster shown in FIG. That is, in this case, since the diagnosis target data does not belong to any normal model range, the diagnosis unit 152e determines that there is a sign of abnormality in the corresponding mechanical equipment 2.

Then, the diagnosis unit 152e stores the presence / absence of an abnormality sign using the diagnosis target data in the diagnosis result storage unit 17 in association with the diagnosis target data.
Incidentally, in the above description, the diagnosis unit 152e determines whether or not there is an abnormality sign depending on whether or not the value of the degree of abnormality u is 1 or more. The value to be is not limited to 1. That is, the diagnosis unit 152e takes the threshold value into consideration in consideration of the distribution (distribution, etc.) of learning target data in a cluster having the cluster center closest to the diagnosis target data, the temporal change rate of the diagnosis target data, Can be set as appropriate.
In the abnormality sign diagnosis apparatus 1 according to the present embodiment, the diagnosis time required for the degree-of-abnormality calculation unit 1152d and the diagnosis unit 152e per machine facility 2 is about 10 seconds.

The contribution degree calculation unit 152f reads out diagnosis target data for each mode from the mode-specific diagnosis target data storage unit 152c, and calculates the contribution degree of each sensor (not shown) to the diagnosis target data. Each of the sensors indicates a plurality of sensors installed in the mechanical equipment 2 shown in FIG.
Although detailed description is omitted, for example, when the characteristic β shown in FIG. 7 is a normalized value of a temperature detected by a specific sensor P (not shown) provided at a predetermined location of the mechanical equipment 2, The contribution degree i is expressed as a ratio of the distance f which is the characteristic β component of the distance d shown in FIG. That is, the contribution degree i is expressed by the following formula (2).
i = f / d Formula (2)

In short, the contribution degree i is a value representing how much the sensor data detected by each sensor contributes to the abnormality degree u. That is, by comparing the value of the contribution degree i for each sensor, when it is detected that there is a sign of abnormality in the specific mechanical equipment 2, which sensor among the plurality of sensors installed in the mechanical equipment 2 is detected. It is possible to specify whether or not contributes most to the detection of an abnormal sign.
The contribution degree calculation means 152f associates the diagnosis target data with the contribution degree i, and stores them in the diagnosis result storage means 17 (see FIG. 3).

<Flow of diagnosis processing by the data mining diagnosis unit>
FIG. 11 is a flowchart showing a diagnosis process in the data mining diagnosis unit. In step S31, the diagnosis target data acquisition unit 152a (see FIG. 3) acquires the diagnosis target data from the sensor data storage unit 12. For example, the diagnosis target data acquisition unit 152a acquires the sensor data for the previous day from the sensor data storage unit 12 as diagnosis target data at a predetermined time (for example, midnight).
In step S32, the mode determination unit 152b determines the operation mode of the diagnosis target data. In step S33, the mode determination unit 152b stores the diagnosis target data for which the operation mode has been determined in the mode-specific diagnosis target data storage unit 152c. Next, in step S34, the abnormality degree calculation unit 152d selects an operation mode to be diagnosed, and acquires diagnosis target data from the mode-specific diagnosis target data storage unit 152c.

In step S35, the abnormality degree calculation unit 152d calculates the abnormality degree of the diagnosis target data. That is, the abnormality level calculation unit 152d acquires a normal model corresponding to the operation mode selected in step S34 from the normal model data storage unit 151e (see FIG. 3), and calculates the abnormality level based on the normal model.
In step S36, the diagnosis unit 152e executes diagnosis. That is, as described above, the diagnosis unit 152e diagnoses the presence or absence of an abnormality sign in the mechanical equipment 2 based on whether or not the value of the degree of abnormality calculated in step S35 is equal to or greater than a predetermined value. In step S37, the diagnosis unit 152e stores the diagnosis result in the diagnosis result storage unit 17. Next, in step S38, the abnormality degree calculation unit 152d determines whether there is another operation mode in which the abnormality degree should be calculated. If there is another operation mode in which the degree of abnormality should be calculated (step S38 → Yes), the processing of the degree-of-abnormality calculation unit 152d returns to step S34. When there is no other operation mode in which the degree of abnormality is to be calculated (step S38 → No), the processing of the data mining diagnosis unit 152 returns.

FIG. 12 is a diagram illustrating an example of a diagnosis result by the data mining diagnosis unit. As shown in FIG. 12, the diagnosis result storage means 17 (see FIG. 3) stores the diagnosis results for each operation mode as a database.
A time 201 shown in FIG. 12 indicates the date and time when the sensor data acquisition unit 11 (see FIG. 1) has acquired a measurement value detected by a sensor (not shown) installed in the machine facility 2. The measurement value is a value obtained by the sensor of the mechanical equipment 2 for each preset period. Therefore, the time 201 is a value obtained by sequentially adding the period (for example, 30 seconds) to a predetermined time (for example, midnight).

  The learning target data acquisition date 202 indicates that a cluster that is a normal model has been learned using the sensor data of this day. The cluster number 203 is an identification number of the cluster having the shortest distance from the cluster center c (see FIG. 7) to the diagnosis target data among the plurality of clusters as the normal model described above. In the abnormality flag 204, a value of 1 is stored when there is a sign of abnormality, and a value of 0 is stored when there is no sign of abnormality, according to the diagnosis result by the diagnostic means 152e (see FIG. 3). The operation state number 205 is an identification number assigned corresponding to the operation mode of the mechanical equipment 2. As described above, the abnormality degree 206 is an abnormality degree calculated by the abnormality degree calculating unit 152d (see FIG. 3). An area 207 shown in FIG. 12 shows an actual measurement value that is a value detected by each sensor installed in the machine facility 2 and a value of a component corresponding to the sensor 1 among the coordinates representing the cluster center c described above. The learning value is stored. The contribution 208 is a contribution calculated for each sensor described above.

<Configuration of remote monitoring unit>
FIG. 13 is a configuration diagram of a remote monitoring unit of the abnormality sign diagnosis apparatus. As described above, the remote monitoring unit 16 sets the upper and lower thresholds for the value (or change amount) of the sensor data acquired from the mechanical equipment 2 based on the experience and knowledge of the manager, When a value or the like exceeds a predetermined range defined by the upper and lower thresholds, it is diagnosed that there is a sign of abnormality.
The remote monitoring unit 16 may be referred to as individual determination means 161 ₁₁ , 161 ₁₂ ,..., 161 _1N , 161 ₂₁ , 161 _{22 ,.} And diagnostic means 162a, 162b,... (Hereinafter, simply referred to as diagnostic means 162). The individual determination means 161 ₁₁ , 161 ₂₁ ,... Acquire sensor data 1 (for example, temperature) from the sensor data storage means 12, and each outputs the determination result to the diagnosis means 162a, 162b,. To do. Similarly, the individual determination means 161 _1N , 161 _2N ,... Acquire the sensor data N from the sensor data storage means 12, and each outputs the determination result to the diagnosis means 162a, 162b,.
Each diagnosis unit 162 is provided with a predetermined logic circuit corresponding to the diagnosis contents (for example, diagnosis of whether or not the cooling water pressure of the mechanical equipment 2 is reduced).

FIG. 14 is an explanatory diagram of diagnosis of an abnormality sign by the individual determination unit of the remote monitoring unit. As described above, each individual determination unit 161 is associated with each sensor installed in the machine facility 2 on a one-to-one basis. As shown in FIG. 14, the individual determination means 161 is used when the measured value of the sensor data is smaller than a preset threshold A (see the one-dot chain line in FIG. 14) or from the threshold B. When it becomes larger (see the broken line in FIG. 14), a predetermined ON signal is output to the diagnostic means 162 connected to itself.
Incidentally, in the individual determination means (for example, 161 ₁₁ and 161 ₂₁ ) for acquiring the same type of sensor data, the threshold values (see A and B in FIG. 14) may be set to be the same, and the diagnostic contents Different threshold values may be set corresponding to.
The individual determination unit 161 performs the determination on the sensor data stored in the sensor data storage unit 12 every preset sampling cycle (for example, 30 seconds), and outputs the result to the diagnosis unit 162. The diagnosis unit 162 has a predetermined logic circuit as described above, and outputs the presence / absence of an abnormality sign corresponding to each diagnosis content corresponding to the presence / absence of a signal from each individual determination unit 161. It is supposed to be.
Then, the diagnosis unit 162 stores each diagnosis result in the diagnosis result storage unit 17 (see FIG. 3).

≪Experimental data≫
FIG. 15 is a diagram illustrating an example of experimental data obtained by the abnormality sign diagnosis apparatus, where (a) is a graph showing changes in measured values, learned values, and contributions, and (b) is a graph showing changes in abnormalities. The experimental data in FIG. 15 shows the case where a gas engine is used as the mechanical equipment 2 and shows the result of learning and diagnosis by the data mining unit 15.
As shown in FIG. 15 (a), before the time t1, the coolant pressure, which is one of the measured values of the gas engine, is maintained in a substantially stable state corresponding to each operation mode. Further, as shown in FIG. 15 (b), the degree of abnormality remains low before time t1. On the other hand, the cooling water pressure gradually decreases after time t1 (see FIG. 15A), and the value of the degree of abnormality increases accordingly (see FIG. 15B).
The data mining diagnosis unit 152 (see FIG. 3) detects an abnormality sign at time t1 when the degree of abnormality exceeds a predetermined threshold value set in advance. That is, it can be seen that the abnormal sign is accurately detected by the abnormal sign diagnostic apparatus 1 before the abnormal situation occurs.

(A) of FIG. 16 is a graph which shows the change of the measured value in the period before and behind performing maintenance work, and the change of the abnormal value in a comparative example, (b) is the abnormality sign diagnostic apparatus concerning this embodiment. It is a graph which shows the change of the abnormal value in the period before and behind performing a maintenance work at the time of using. In addition, the experimental data of FIG. 16 shows the case where a gas engine is used as the mechanical equipment 2, and shows the result of learning and diagnosis by the data mining unit 15.
As shown in the upper diagram of FIG. 16A, for example, when maintenance such as increasing the amount of cooling water is performed, the state of the mechanical equipment 2 changes significantly due to the maintenance. Even when such maintenance is performed on the mechanical equipment 2, the learning data acquisition unit 151a (see FIG. 3) temporarily sets the learning target data as sensor data in the normal mode (see FIG. 4) described above. When the information is acquired from the storage unit 12, the following situation occurs. That is, although the mechanical equipment 2 is in a normal state due to maintenance, the diagnostic means 152e (see FIG. 3) has a sign of abnormality in the mechanical equipment 2 after the maintenance as shown in the lower diagram of FIG. It will continue to be judged.

  On the other hand, in the abnormality sign diagnosis apparatus 1 according to the present embodiment, the learning target data acquisition unit 151a (see FIG. 3) learns in the maintenance mode (see FIG. 5) according to the maintenance information input to the computer 3 (see FIG. 1). Select a period. That is, as shown in FIG. 5, the learning target data acquisition unit 151a secures a predetermined learning period B1 after maintenance, and restarts the diagnosis after the learning period B1 has elapsed. Accordingly, as shown in FIG. 16 (b), the diagnosis means 152e (see FIG. 3) sets the mechanical equipment 2 for which the learning period after maintenance has passed to normal (that is, no abnormal signs), and provides an appropriate diagnosis result. You can see that it is outputting.

Incidentally, as described with reference to FIG. 5, the learning unit 151d cannot generate an appropriate normal model during the maintenance period and the minimum necessary learning period B1 after the maintenance. (See FIG. 1) does not output the diagnosis result.
However, the remote monitoring unit 16 (see FIG. 1) performs diagnosis based on sensor data input every sampling period (for example, 30 seconds), and diagnoses the diagnosis result storage means 17 (see FIG. 1). The result is output. That is, even during the maintenance period and the minimum learning period B1 required after the maintenance, the abnormality sign diagnosis device 1 outputs the diagnosis result by the remote monitoring unit 16 and displays it on the display means 19.

As described above, the setting of the threshold in the individual determination unit 161 of the remote monitoring unit 16 shown in FIG. 13 and the setting of the logic circuit in the diagnosis unit 162 can be changed as appropriate. Therefore, for example, the administrator of the abnormality sign diagnosis apparatus 1 can appropriately change the setting of the threshold value or the like with reference to the maintenance period and the maintenance content.
Incidentally, as shown in FIG. 1, the display control unit 18 stores the sensor data stored in the sensor data storage unit 12 and the maintenance information storage in addition to the diagnosis result by the data mining unit 15 and the diagnosis result by the remote monitoring unit 16. Maintenance information stored in the means 14 can also be displayed on the display means 19.

≪Example of screen display by display control means≫
FIG. 17 is a diagram illustrating an example of a case where the diagnosis result by the abnormality sign diagnosis apparatus is displayed on the display unit. First, the legend 400 shown in FIG. 17 will be described. The legend 401 “During operation / with alarm” indicates that the machine facility 2 is in operation on the diagnosis date, and that the alarm by the interlock (not shown) of the machine facility 2 described above has occurred. . The legend 402 “During operation / no alarm” indicates that the machine facility 2 is in operation on the diagnosis date and no alarm is generated by the interlock.
Legend 403 “During operation / with alarm (low importance)” indicates that the machine facility 2 is in operation on the diagnosis date and an alarm is generated by an interlock (not shown). Indicates a low-severity alarm with a low probability of failure. The legend 404 “During operation / no warning” indicates that the machine facility 2 is in operation on the date of diagnosis and no interlock alarm has occurred, but the data mining unit 15 (see FIG. 1) Indicates that there was a sign of abnormality.

A legend 405 “stopped / alarmed” indicates that the machine facility 2 is stopped on the diagnosis date and an alarm by an interlock (not shown) is generated. The legend 406 “stopped / no alarm” indicates that the machine facility 2 is stopped on the diagnosis date and no interlock alarm has occurred.
A legend 407 “in maintenance” indicates that maintenance was performed on the machine facility 2 on the diagnosis date. This corresponds to, for example, the maintenance period (3 to 5 days) shown in FIG. A legend 408 “no data” indicates that the learning target data acquisition unit 151a (see FIG. 3) could not acquire sufficient learning target data for some reason.

A legend 409 “After maintenance / with alarm (not subject to learning)” indicates that the alarm by the interlock described above occurred during the learning period immediately after the maintenance of the mechanical equipment 2 was performed. Since there is a possibility that the state of the machine facility 2 is not normal, the sensor data in this period is excluded from the learning target data (for example, see “with alarm (exclusion)” in FIG. 4). The legend 410 “After maintenance / no alarm (during learning)” indicates that the diagnosis date is included in the learning period immediately after the maintenance of the mechanical equipment 2 is performed. This corresponds to, for example, the learning period B1 shown in FIG.
The legend 411 “No sign” indicates that the machine facility 2 was diagnosed as having no abnormal sign (normal) by the data mining unit 15 (see FIG. 1) on the diagnosis date. The legend 412 “with a sign” indicates that the machine facility 2 has been diagnosed as having a sign of abnormality by the data mining unit 15 (see FIG. 1) on the diagnosis date. A legend 413 “data present / undiagnosed” indicates that learning target data sufficient to learn a normal model could not be acquired and the presence / absence of an abnormality sign was not diagnosed. That is, when the sensor data acquired from the sensor data storage unit 12 by the learning target data acquisition unit 151a (see FIG. 3) is insufficient to generate a normal model, the data mining unit 15 (see FIG. 1) A signal indicating the state is output to the display control means 18.

  The “date” 421 indicates the date of the day when the diagnosis is performed on each machine facility 2, and by scrolling the scroll bar 422, the diagnosis result for a predetermined period can be displayed. . The “registered site name” 423 is a customer site name that manages at least one machine facility 2. “Unit No.” 424 is an identification number for specifying the machine equipment 2 managed by each registered site. By scrolling the scroll bar 425, the diagnosis results of all registered machine equipment 2 are displayed. It can be displayed.

  For example, the registration number “No. 2” of the registered site “site 2” displayed second from the top of the registered site name 423 is displayed. A description will be given by taking the “01” mechanical equipment 2 as an example. In the machine facility 2, it is understood that maintenance was performed on February 22, 2010, although there was no alarm on February 20, 2010 and 21st. As described above with reference to FIG. 5, when maintenance is performed, sensor data before that (before February 21, 2010) is not used as learning target data of the data mining unit 15 (see FIG. 1). Then, the data mining unit 15 sets the learning period B1 (see FIG. 5) after maintenance from February 23 to March 1, 2010 after maintenance.

Incidentally, as described with reference to FIG. 5 above, the period from the maintenance to the end of the learning period B1 (February 22, 2010 to March 1, 2010) is the data mining unit. 15 does not diagnose the presence or absence of abnormal signs.
In addition, as shown in FIG. 17, during the period from March 11 to 13, 2010 and from March 15 to 19, 2010, the data mining unit 15 is diagnosed as having an abnormal sign. As a result, the registration number “Site 2” No. It can be seen that the machine equipment 2 of “01” has a high possibility of occurrence of an abnormality (failure or the like) in the near future. Therefore, the manager who has seen the display screen shown in FIG. 17 can pay attention to the trend of the machine facility 2 and analyze the cause of the abnormal sign to take future countermeasures.

In the example of the display screen of FIG. 17 described above, the example in which the display control unit 18 (see FIG. 1) of the abnormality sign diagnosis apparatus 1 displays the diagnosis result by the data mining unit 15 on the display unit 19 has been described. On the other hand, as described above, the diagnosis by the remote monitoring unit 16 (see FIG. 1) is also made every predetermined period (for example, 30 seconds).
FIG. 18 is a diagram illustrating an example of a case where a diagnosis result by the remote monitoring unit is displayed on the display unit. FIG. 18A illustrates an abnormal sign detection screen when an abnormal sign is detected, and FIG. 18B illustrates an abnormal sign. It is a figure which shows the example of the abnormal sign detail screen which displays the content of.
For example, when there is a machine facility 2 diagnosed by the remote monitoring unit 16 as having an abnormal sign, the abnormal sign detection screen shown in FIG. 18A is popped up on the screen shown in FIG. 17 described above. The On the abnormality sign detection screen, as indicated by reference numeral 501, a message indicating that the abnormality sign has been detected by the remote monitoring unit 16 pops up on the display means 19. The “abnormality sign detection site” 502 displays a site where the abnormal sign is detected (the name of the site that manages the mechanical equipment 2).

Further, when the administrator who has seen the abnormality sign detection screen clicks on the detailed display 503 shown in FIG. 18A, the display control means 18 displays the abnormality sign detail screen shown in FIG. Display. The abnormality sign detail screen includes, for example, “check date and time” 504 diagnosed by the remote monitoring unit 16 as having an abnormality sign, “management No.” 505 for specifying the mechanical equipment 2, and details of abnormality sign detection by the remote monitoring unit 16. “Abnormal sign name” 506 indicating “abnormal sign”, “abnormal level” 507 indicating the degree of abnormality sign, “occurrence time” 508 which is the time when the abnormal sign starts to be detected, and abnormal signs detected until then are detected The “end time” 509, which is the time when it disappeared, is displayed.
Incidentally, the “abnormal level” 507 is a predetermined numerical level (for example, 1, 2, 3, etc.) depending on how much the sensor data deviates from the normal range defined by the threshold value described above (see FIG. 14). ).

As described above, the display control unit 18 causes the display unit 15 to display diagnosis results and maintenance information from the data mining unit 15 (see FIG. 1) (see FIG. 17), and the remote monitoring unit 16 (see FIG. 1). ), A screen showing the diagnosis result is displayed in a pop-up on the display means 19.
And the administrator who grasped the above-mentioned diagnosis result can also display sensor data of machine equipment 2 diagnosed as having a sign of abnormality, for example, specifying a predetermined period.

FIG. 19 is a diagram illustrating an example of a case where changes in measurement values by a plurality of sensors installed in mechanical equipment are displayed on a display unit.
For example, the administrator who has viewed the display screen of FIG. 17 has received information about the machine equipment 2 of the machine number “01” of the site name “site 1” displayed at the top of the registered site name 423 from February 23, 2010. Consider the case of looking at sensor data on the 25th. In this case, when the above site name and machine number are specified and a predetermined graph display button (not shown) is clicked, a graph screen shown in FIG. 19 is displayed on the display means 19 (see FIG. 1).

  As shown in FIG. 19, when the administrator inputs a predetermined period in the field indicated by 601, the display control unit 18 stores the sensor data of the mechanical equipment 2 in the period 2 in the sensor data storage unit 11 (see FIG. 1). Get from. Then, the display control unit 18 causes the display unit 19 to display a graph of the temporal change of each sensor data during the period. The display control means 18 causes the display means 19 to display which sensor data each graph displays as a legend 603. Further, when the administrator designates a specific time (for example, time t0) in the graph 602 by using an input unit (not shown) such as a cursor, the display control unit 18 sets the value of each sensor data at the time to 604. Can be displayed in the display column.

≪Display processing by display control means≫
FIG. 20 is a flowchart showing display processing by the display control means. In step S41, the display control means 18 (see FIG. 1) determines whether or not the diagnosis result of the abnormal sign by the data mining unit 15 (see FIG. 1) is obtained in the diagnosis result storage means 17 (see FIG. 1). .
When the diagnosis result of the abnormal sign by the data mining unit 15 is obtained (step S41 → Yes), the processing of the display control unit 18 proceeds to step S42. On the other hand, when the diagnosis result of the abnormal sign by the data mining unit 15 is not obtained (step S41 → No), the processing of the display control unit 18 proceeds to step S45.
In step S42, the display control means 18 determines whether or not the diagnosis result by the data mining unit 15 is “abnormal sign”. When the diagnosis result by the data mining unit 15 is “abnormal sign” (step S42 → Yes), the display control unit 18 causes the display unit 19 to display a display corresponding to “abnormal sign” (see legend 412 in FIG. 17). ). On the other hand, when the diagnosis result by the data mining unit 15 is “no abnormality sign” (step S42 → No), the display control means 18 causes the display means 19 to display a display corresponding to “no abnormality sign” (the legend of FIG. 17). 411).

  In step S45, the display control means 18 determines whether or not the diagnosis date is within the maintenance period. Incidentally, the information is stored in the diagnosis result storage means 17 by the learning target data acquisition means 151a (see FIG. 3) (see step S24 in FIG. 10). If the diagnosis date is within the maintenance period (step S45 → Yes), the display control means 18 causes the display means 19 to display a display corresponding to “in maintenance” (see legend 407 in FIG. 17). On the other hand, if the diagnosis date is not within the maintenance period (step S45 → No), the process of the display control means 18 proceeds to step S47.

  In step S47, the display control means 18 determines whether or not the diagnosis date is included in the learning period B1 (see FIG. 5). Incidentally, the information is stored in the diagnosis result storage unit 17 by the learning target data acquisition unit 151a (see FIG. 3) (see step S26 in FIG. 10). When the diagnosis date is included in the learning period B1 (step S47 → Yes), the display control unit 18 causes the display unit 19 to display a display corresponding to “during the learning period after maintenance” (see the legend 410 in FIG. 17). When the diagnosis date is not included in the learning period B1 (step S47 → No), the process of the display control unit 18 proceeds to step S42.

  On the other hand, as described above, the remote monitoring unit 16 (see FIG. 1) diagnoses the presence / absence of abnormality of the mechanical equipment 2 at predetermined intervals, and the diagnosis result is stored in the diagnosis result storage unit 17 (see FIG. 1). ). Therefore, the display control means 18 causes the display means 19 to display the diagnosis result from the remote monitoring section 16 independently of the diagnosis result from the data mining section 15 (step S49 in FIG. 20). For example, when the remote monitoring unit 16 diagnoses that there is a sign of abnormality, the display control unit 18 pops up the diagnosis result on the display unit 19 as described above (see FIG. 18).

<< Effect of the present invention >>
As described above, in the abnormality sign diagnosis apparatus 1 according to the present embodiment, the learning target data acquisition unit 151a stores the learning target data in accordance with the maintenance information received from the external computer 3 via the communication network N. You can specify the period to get.
For example, if the diagnosis date is within the normal period described above, the learning target data acquisition unit 151a selects the sensor data of the learning period B that goes back in the past by the predetermined period A (see FIG. 4) as the learning target period. The predetermined period A and the learning period B can be appropriately changed, and an appropriate period can be set as the learning period.

In addition, when the diagnosis date is included in the maintenance period (see FIG. 5) or the subsequent learning period B1, the learning object data acquisition unit 151a does not acquire learning object data from the sensor data storage unit 12. Therefore, the abnormality sign diagnosis apparatus 1 can prevent the learning unit 151d from performing useless learning using learning target data that is inappropriate or lacks the number of data.
Then, the learning target data acquisition unit 151a causes the diagnosis result storage unit 17 to store information indicating that the diagnosis date is included in the maintenance period or the subsequent learning period B1. Therefore, the abnormality sign diagnosis apparatus 1 can display the maintenance period and the subsequent learning period B1 on the display unit 19.

Further, when maintenance is performed on the machine facility 2, the learning target data acquisition unit 151a acquires sensor data after the maintenance is completed and the state of the machine facility 2 is stabilized as learning target data. Therefore, the abnormality sign diagnosis apparatus 1 can appropriately determine the presence or absence of an abnormality sign even after maintenance.
Further, during the maintenance period and the learning period B1 (see FIG. 5), the data mining unit 15 (see FIG. 1) does not diagnose the presence / absence of an abnormality sign, but the remote monitoring unit 16 (see FIG. 1) always A diagnosis result is output at every predetermined sampling period. Therefore, the abnormality sign diagnosis apparatus 1 can diagnose the presence or absence of an abnormality sign even in the maintenance period and the learning period B1.

The setting of the threshold value in the individual determination unit 161 of the remote monitoring unit 16 (see FIG. 13) and the setting of the logic circuit in the diagnosis unit 162 can be changed as appropriate. Therefore, for example, the administrator of the abnormality sign diagnosis apparatus 1 can appropriately change the setting of the threshold value or the like with reference to the maintenance period and the maintenance content. That is, the abnormality sign diagnosis apparatus 1 can appropriately output the presence or absence of an abnormality sign even during the maintenance period and the learning period B1.
Furthermore, the abnormality sign diagnosis apparatus 1 can diagnose the presence or absence of an abnormality sign independently by the data mining unit 15 and the remote monitoring unit 16 and display each diagnosis result on the display unit 19. Therefore, the administrator who sees the display result can analyze the cause of the abnormal sign in consideration of both the diagnosis result by the data mining unit 15 and the diagnosis result by the remote monitoring unit 16, and can respond quickly. It becomes.

In general, since the data mining unit 15 is highly sensitive, it is possible to detect the presence or absence of an abnormality sign earlier than the remote monitoring unit 16. On the other hand, since the remote monitoring unit 16 has a relatively simple correspondence between the sensor data and the diagnosis result (for example, a one-to-one correspondence), it is relatively easy to identify the cause of the abnormality sign. It is.
Therefore, since the abnormality sign diagnosis apparatus 1 takes advantage of both the advantages of the data mining unit 15 and the remote monitoring unit 16, the abnormality sign can be detected at an early stage and the cause can be accurately analyzed. It becomes.

Moreover, according to the abnormality sign diagnostic apparatus 1 according to the present embodiment, as shown in FIG. 17, the maintenance period of each registered machine facility 2, the learning period necessary after the maintenance, and the end of the learning period The diagnosis period until the next maintenance is performed can be displayed in an identifiable manner for each date. In addition, the presence or absence of an abnormality sign of each machine facility during the diagnosis period can be displayed in an identifiable manner. Therefore, the administrator of the abnormality sign diagnostic apparatus 1 can carefully monitor the mechanical equipment 2 diagnosed as having an abnormal sign, and can examine the response to the mechanical equipment 2 at an early stage.
Further, according to the abnormality sign diagnosis apparatus 1 according to the present embodiment, the state of each mechanical facility 2 for each date can be displayed on the display means 19, and the administrator can grasp the state at a glance. In general, as the number of machine facilities 2 registered in the abnormality sign diagnosis apparatus 1 increases, it becomes more difficult to grasp the relationship. According to the abnormality sign diagnosis apparatus 1, even when there are many registered machine equipments 2, the administrator can easily grasp the state of each date of the machine equipment 2 and more appropriately each machine equipment 2 2 management can be performed.

Further, the abnormality sign diagnosis device 1 can display the diagnosis result by the remote monitoring unit 16 on the display means 19 in addition to the diagnosis result by the data mining unit 15 (see FIG. 18). As described above, since the correspondence between the sensor data and the diagnosis result is relatively simple in the remote monitoring unit 16, it is relatively easy to specify the cause of the abnormality sign. Furthermore, as shown in FIG. 19, the abnormality sign diagnosis apparatus 1 can display each sensor data within a predetermined period in the mechanical equipment 2 specified by the administrator as a graph.
For example, the administrator of the abnormality sign diagnosis apparatus 1 can discover the machine facility 2 having an abnormality sign early by diagnosis by the data mining unit 15 and carefully monitor the machine facility 2 (see FIG. 17). Then, by using the diagnosis result by the remote monitoring unit (see FIG. 18) and the sensor data of the machine facility 2 (see FIG. 19), the cause of the abnormal sign in the machine facility 2 can be analyzed appropriately.

≪Modification≫
In the above description, the sensor data and maintenance information of the mechanical equipment 2 are transmitted to the abnormality sign diagnosis apparatus 1 via the communication network N, but the present invention is not limited to this. For example, the abnormality sign diagnosis apparatus 1 can acquire each data from a storage medium storing the sensor data or maintenance information.
In the above description, the mode determination unit 151b (see FIG. 3) determines the operation mode based on the value (or rate of change) of the learning target data, but the present invention is not limited to this. That is, the mechanical equipment 2 may detect its own operation mode, add information on the operation mode to the sensor data, and transmit the information to the abnormality sign diagnosis apparatus 1 via the communication network N. In this case, the mode determination unit 151b refers to the operation mode added to the sensor data, and stores the learning target data in the mode-specific learning target data storage unit 151c for each operation mode.

Moreover, although FIG. 1 demonstrated the case where the mechanical installation 2 and the computer 3 existed in different positions and the maintenance information of each mechanical installation 2 was registered into one computer 3 collectively, it is not restricted to this. Absent. That is, there are a plurality of computers 3, each computer 3 manages maintenance information of one or a plurality of (for example, five) machine facilities 2, and the maintenance information is transmitted to the abnormality sign diagnosis apparatus 1 via the communication network N. It is good also as transmitting.
In the above, the data mining diagnosis unit 152 (see FIG. 3) uses the sensor data for one day of the previous day as diagnosis target data once a day at a preset time (for example, midnight). Although acquired from the sensor data storage means 12, it is not restricted to this. For example, the data mining diagnosis unit 152 may diagnose an abnormal sign in real time.

In the above description, the contribution calculated by the contribution calculation unit 152f of the data mining diagnosis unit 152 is simply stored in the diagnosis result storage unit 17 and further displayed on the display unit 19. However, the present invention is not limited to this. . As described above, it is possible to grasp how much the measurement value by the sensor contributes to the degree of abnormality based on the contribution calculated for each sensor (not shown) installed in the mechanical equipment 2.
Therefore, the contribution calculated by the contribution calculation unit 152f may be reflected on the logic circuit of the diagnosis unit 162 of the remote monitoring unit 16 (see FIG. 13). For example, let us consider a case where the data mining unit 15 finds that the contribution degree of the sensor A is the largest for the content of a certain abnormality sign. In this case, if the logic circuit of the diagnostic unit 162 is reconfigured so that the change in the value of the sensor A is most sensitively reflected in the diagnostic result in the remote monitoring unit 16, the remote monitoring unit 16 can more appropriately display the diagnostic result. It becomes possible to output.

In the above description, the mode determination units 151b and 152b are configured to determine the operation mode of the mechanical equipment 2. However, the present invention is not limited to this. That is, in the learning process and the diagnosis process in the data mining unit 15, when it is not necessary to determine the operation mode of the machine facility 2, the operation mode determination process by the mode determination units 151b and 152b can be omitted.
In addition, as shown in FIG. 3, the mode-specific learning target data storage unit 151c, the normal model data storage unit 151e, and the mode-specific diagnosis target data storage unit 152c are included in the data mining unit 15, but the present invention is not limited thereto. That is, each of the storage means may be provided outside the abnormality sign diagnosis apparatus 1.
In the above description, the case where the learning unit 151d performs clustering using the k-average method as non-hierarchical clustering to learn a normal model is described, but the present invention is not limited thereto. That is, for example, the learning unit 151d may perform learning processing using fuzzy clustering, a mixed density distribution method, or the like as non-hierarchical clustering.

In the example of the display screen shown in FIG. 17, the display control means 18 displays the diagnosis result of the data mining unit 15 (see FIG. 1) as the presence / absence of a sign (see legends 411 and 412). Not limited to. That is, when either one or both of the data mining unit 15 and the remote monitoring unit 16 diagnoses that there is an abnormal sign in the machine facility 2, the display control unit 18 displays the legend 412 on the display unit 19 (see FIG. 17). It is also possible to display the “predictive sign” shown in FIG. Further, in this case, the display control means 18 may display the diagnosis result output from the remote monitoring unit 16 on the display means 19 during the maintenance of the mechanical equipment 2 or during a predetermined learning period immediately after that.
Further, in the above case, the display control means 18 is configured such that only the data mining unit 15 diagnoses that there is an abnormal sign, only the remote monitoring unit 16 diagnoses that there is an abnormal sign, and the data mining unit 15 and the remote monitoring unit. It is good also as making it display on the display means 19 so that identification of the case where both 16 may diagnose abnormality is present.

  In the example of the display screen shown in FIG. 18, the display control unit 18 displays the diagnosis result by the remote monitoring unit 16, but is not limited thereto. For example, the display control means 18 may cause the display means 19 to display the presence or absence of an alarm by an interlock (not shown) installed in the machine facility 2 in the same format as the display screen shown in FIG.

Further, in the above, the display control means 18 displays the diagnosis result by the data mining unit 15 (see FIG. 17), the diagnosis result by the remote monitoring unit 16 (see FIG. 18), and the sensor data (see FIG. 19) in the specific mechanical equipment 2. Is displayed on one display means 19, but is not limited thereto.
That is, there may be a plurality of display means 19 and the display control means 18 may display each diagnosis result on each display means 19.

DESCRIPTION OF SYMBOLS 1 Abnormal sign diagnostic apparatus 10 Communication means 11 Sensor data acquisition means 12 Sensor data storage means 13 Maintenance information acquisition means (control means)
14 Maintenance information storage means 15 Data mining section (control means)
151 Data mining learning unit 151a Learning object data acquisition means (control means)
151b Mode determination means 151c Mode-specific learning target data storage means 151d Learning means (control means)
151e Normal model data storage unit 152 Data mining diagnosis unit 152a Diagnosis target data acquisition unit (control unit)
152b Mode determination means 152c Mode-specific diagnosis object data storage means 152d Abnormality calculation means (control means)
152e Diagnosis means (control means)
152f Contribution calculation means 16 Remote monitoring unit (control means)
161, 161a, 161b, ..., 161n Individual determination means (control means)
162 Diagnosis means (control means)
17 Diagnosis result storage means 18 Display control means (control means)
19 Display means 2 Machine equipment 3 Computer N Communication network

Claims (7)

In a method for displaying an abnormality sign diagnosis result by an abnormality sign diagnosis device for diagnosing the presence or absence of an abnormality sign of the machine equipment using sensor data detected by various sensors installed in at least one machine equipment,
The abnormal sign diagnostic apparatus is
Sensor data storage means for storing sensor data from the various sensors acquired at predetermined sampling periods;
Maintenance information storage means for storing maintenance information of the mechanical equipment;
Control means,
The control means includes
Corresponding to maintenance information acquisition procedure for acquiring maintenance information from the maintenance information storage means including at least information specifying machine equipment and maintenance work period in which maintenance work is performed, and maintenance information acquired by the maintenance information acquisition procedure A learning target data acquisition procedure for designating a period during which the sensor data was acquired and reading out sensor data corresponding to the period from the sensor data storage means, and learning for learning a normal model indicating a normal range of the sensor data A first diagnostic step comprising: a procedure; and a diagnostic procedure for diagnosing the presence or absence of an abnormality sign of the mechanical equipment based on the normal model;
A second diagnostic step of reading out the sensor data from the sensor data storage means and diagnosing that there is a sign of abnormality when each sensor data exceeds a predetermined threshold value set in advance;
A display control step for performing display control of information including the diagnosis result of the first diagnosis step and the diagnosis result of the second diagnosis step;
In the display control step, at least a period during which the maintenance work is performed, a predetermined learning period after the maintenance work, a diagnosis period after the predetermined learning period until the next maintenance work is performed, and the diagnosis period A display method of an abnormality sign diagnosis result, wherein the presence or absence of an abnormality sign of the mechanical equipment is displayed on a display means so as to be distinguishable. The method of displaying an abnormality sign diagnosis result according to claim 1, wherein the display control step displays on the display means whether or not there is an abnormality sign of each of the mechanical equipment on each diagnosis date. The display control step causes the display means to display the presence / absence of an abnormality sign of each machine facility in the diagnosis period as a diagnosis result of the first diagnosis step, and whether or not a diagnosis date is the diagnosis period Regardless of the above, the abnormality predictor diagnosis result display method according to claim 2, wherein the display means displays the diagnosis result of each of the machine facilities in the second diagnosis step. 4. The abnormality sign diagnosis result according to claim 3, wherein, when the display control step is diagnosed as having an abnormality sign by the second diagnosis step, the diagnosis result is pop-up displayed on the display unit. 5. How to display. The said display control process is further made to display on the said display means so that identification of the driving | running state of each said mechanical installation on each diagnosis date is possible. The Claim 1 characterized by the above-mentioned. Display method of abnormal sign diagnosis results. The display control step further displays on the display means in an identifiable manner whether or not there is an alarm by an alarm device installed in each of the machine equipment itself on each diagnosis date. The display method of the abnormal sign diagnostic result as described in any one of Claims 5-6. In the display control step, it is further determined that the sensor data that can be acquired from the sensor data storage means is insufficient to learn the normal model when diagnosing each of the mechanical facilities on each diagnosis date. 7. If it is made in the first diagnosis step, the display means displays that the presence / absence of an abnormality sign of the mechanical equipment has not been diagnosed. Display method of the abnormal sign diagnosis result described in the item.

Images