JP7418157B2 - Equipment monitoring device and equipment monitoring method - Google Patents

Description

The present disclosure relates to an equipment monitoring device and an equipment monitoring method, and particularly relates to an equipment monitoring device and an equipment monitoring method that determine the health of equipment based on a large amount of measurement data collected from the equipment.

BACKGROUND ART In facilities such as buildings and plants, various equipment such as lighting and air conditioning are installed. Facility managers perform appropriate operations by acquiring and monitoring measurement data regarding the equipment installed in the facility, either periodically or on an as-needed basis.

Conventionally, managers detect abnormalities in equipment by acquiring measurement data that deviates from a preset normal range. Patent Document 1 discloses a technique for determining that equipment is in an abnormal state when a difference between a learning model indicating normal operation and measured data exceeds a predetermined threshold.

Japanese Patent Application Publication No. 2018-13914

However, there are abnormalities or signs of abnormalities that cannot be detected by the technique disclosed in Patent Document 1. For example, when a meter malfunctions, the absolute value of the measured data of the meter is within the normal range, but the value of the measured data may change in a relatively short period.

Therefore, an object of the present disclosure is to provide an equipment monitoring device and an equipment monitoring method that can detect abnormalities or signs of abnormalities that could not be detected conventionally by understanding changes in measurement data.

The equipment monitoring device of the present disclosure includes a data collection unit that collects updated measurement data updated at a management point, a status change history that records changes in measurement data before updating at the management point, and a determination of the health of the management point. an information storage unit that stores a group of normal operation patterns for updating, and a status change unit that adds updated measurement data to the status change history to create an updated status change history when the updated measurement data differs from the status change history. A recording unit, a health determination unit that determines the health of a management point by comparing the status change history after the update with a normal operation pattern group, and a health determination unit that determines the health of the management point, and the status change history after the health determination unit updates the normal operation pattern group. The present invention is characterized by comprising an alarm management unit that notifies an administrator of an alarm when it is determined that the information does not correspond to the group.

According to the equipment monitoring device with the above configuration, the health of the management point is determined by comparing changes in measurement data at the management point with a group of normal operation patterns, and the status change history after update is a pattern of normal operation patterns. You can notify the administrator that the group is not supported. Normal operation patterns include not only the absolute value of measured data but also the rate of change and frequency of change in measured data, so they could not be detected using methods that judge the health of management points using only the absolute value of measured data. Anomalies or signs of anomalies can be detected.

In the equipment monitoring device of the present disclosure, the information storage unit may further store an alarm history in which alarms are recorded. According to this, the manager who has been notified of the alarm can use the alarm history to determine whether there is an abnormality in the equipment.

In the equipment monitoring device of the present disclosure, when the administrator determines that the management point is in a normal state, a new normal operation pattern is created by replacing at least part of the normal operation pattern group with at least part of the updated status change history. The image forming apparatus may further include a normal operation pattern creation unit that creates a new normal operation pattern group by creating a new normal operation pattern group or adding at least a part of the updated state change history to the normal operation pattern group. According to this, it is possible to customize each facility by updating the normal operation pattern group, so it is possible to create a normal operation pattern group suitable for each facility.

In the equipment monitoring device of the present disclosure, the normal operation pattern of one management point included in the normal operation pattern group includes measurement data before updating of one or more other management points or the one management point. It may also include data related to the operation of the equipment. According to this, by including measurement data of other management points that may affect the measurement data of a management point and data related to equipment operation, management points can be measured after considering the correlation with other data. The soundness can be determined.

The equipment monitoring method of the present disclosure includes a data collection step of collecting updated measurement data updated at a management point, and a step in which the updated measurement data differs from a state change history that records changes in measurement data before updating at the management point. includes a state change history creation step that adds updated measurement data to the state change history to create an updated state change history, and a normal operation pattern for determining the state change history after the update and the health of the management point. In the health determination step, which determines the health of the management point by comparing the two groups, if it is determined in the health determination step that the updated state change history does not correspond to the normal operation pattern group, the administrator The method is characterized by including an alarm notification step of notifying an alarm to.

According to the equipment monitoring method with the above configuration, when the measured data at the management point changes, the correspondence with the normal operation pattern including the rate of change and frequency of change of the measured value is checked, so only the absolute value of the measured value is checked. It is possible to detect abnormalities or signs of abnormalities that could not be detected using conventional methods of determining the health of management points.

In the equipment monitoring method of the present disclosure, if the administrator determines that the management point is in a normal state and further determines that the updated state change history should be registered as a normal operation pattern, By creating a new normal operation pattern group by replacing at least part of it with at least a part of the updated status change history, or by adding at least a part of the updated status change history to the normal operation pattern group. The method may further include a normal operation pattern creation step of creating a new normal operation pattern group. According to this, since the normal operation pattern group can be customized for each facility, it is possible to create a normal operation pattern group adapted to each facility.

According to the equipment monitoring device and the equipment monitoring method of the present disclosure, it is possible to detect abnormalities or signs of abnormalities that could not be detected conventionally.

It is a diagram showing the overall configuration of an equipment monitoring device and equipment installed in a facility in an example of an embodiment. FIG. 2 is a hardware configuration diagram of an equipment monitoring device in an example of an embodiment. It is a figure which shows an example of state change history. FIG. 3 is a diagram illustrating an example of a group of normal operation patterns. 5 is a diagram corresponding to FIG. 4 showing another example of a normal operation pattern group including measurement data of other management points. FIG. 7 is a flowchart illustrating processing when a data collection unit collects updated measurement data in an example of the embodiment.

Hereinafter, an example of an embodiment of an equipment monitoring device according to the present disclosure will be described in detail with reference to the drawings. FIG. 1 is a diagram showing the overall configuration of an equipment monitoring device 1 and equipment 3 installed in a facility 2 in an example of an embodiment. The equipment monitoring device 1 collects updated measurement data N from the management points 4 of the equipment 3 installed in the facility 2, such as lighting 3a, air conditioner 3b, electric power meter 3c, entrance/exit control door 3d, etc., and updates the data of each management point 4. Determine the soundness.

The facility 2 is, for example, a building, a plant, or the like. The scale of the facility 2 is also not particularly limited, and may be of any type, such as a large shopping mall or a small building.

In addition to the lighting 3a, air conditioner 3b, power meter 3c, and entrance/exit control door 3d shown in FIG. 1, the equipment 3 includes elevators, escalators, etc. installed in the facility 2, water and gas meters, etc. of the facility 2. can be exemplified.

The management points 4 can be provided for each setting, status, measured value, etc. that should be monitored for the equipment 3. Here, the setting is a condition that defines the operation of the equipment 3, the state is the actual state of the equipment 3, and the measured value is the result of measuring the state around the equipment 3 with a sensor or the like. For example, the setting and state of the lighting 3a can be expressed as ON or OFF, and the setting and state of the air conditioning temperature can be expressed as, for example, 20°C. Moreover, one or more management points 4 can be provided in one facility 3. For example, in the case of the air conditioner 3b, the actual measured temperature, set temperature, actual measured humidity, etc. may be used as the management points 4. Furthermore, one management point 4 can be provided for a plurality of facilities 3. For example, when a plurality of lights 3a are synchronized, one management point 4 may be provided for the plurality of lights 3a.

The updated measurement data N acquired at the management point 4 includes discrete values indicated by ON/OFF indicating the setting and status of the power supply of the lighting 3a, the supply voltage to the facility 2 of 6000V, the set temperature of the air conditioner 3b of 20°C, etc. There is a continuous value indicated by . As long as the management point 4 can be appropriately monitored, the timing of collecting the updated measurement data N is not particularly limited. Depending on the type of measurement data, the management point 4 can automatically send updated measurement data N when there is a change in measurement data, or the management point 4 can send updated measurement data N at regular intervals. You may also do something like this. When collecting updated measurement data N regularly, the time interval for collecting updated measurement data N can be set to, for example, 1 second, 30 seconds, 1 minute, 10 minutes, 1 hour, etc.

The network 5 broadly includes communication networks that can transmit and receive data. The network 5 may be wired or wireless. Further, the network 5 may be either an intranet or the Internet, but the Internet is preferable because the location where the facility monitoring device 1 is installed is not limited.

The administrator 6 determines whether the management point 4 is in a normal state or an abnormal state by interlocking with the equipment monitoring device 1. In addition, the administrator 6 updates and creates a normal operation pattern group 111, which will be described later, by feeding back the determination to the equipment monitoring device 1. The administrator 6 may be a computer instead of a human. If the administrator 6 is a computer, it is preferably a computer with a learning function such as AI (Artificial Intelligence). Further, if the administrator 6 is a human being, the administrator 6 may be one person or may be a collection of two or more persons.

The equipment monitoring device 1 includes a data collection section 10 and an information storage section 11. The data collection unit 10 collects updated measurement data N at the management point 4 via the network 5. Further, the data collection unit 10 may have a function of transmitting a transmission command to cause the management point 4 to acquire measurement data and transmit the measurement data as updated measurement data N.

The information storage unit 11 stores a state change history 110 and a normal operation pattern group 111. The state change history 110 is data that records changes in measurement data at the management point 4 before updating. The normal operation pattern group 111 is a set of normal operation patterns that should be determined to indicate that the management point 4 is healthy with respect to changes in the updated measurement data N. Further, the information storage unit 11 may store an alarm history 112 that records alarms notified to the administrator 6, which will be described later.

Furthermore, the equipment monitoring device 1 includes a state change recording section 12, a health determination section 13, and an alarm management section 14. Furthermore, the equipment monitoring device 1 may include a normal operation pattern creation unit 15 that creates a new normal operation pattern group 111 in response to a request from the administrator 6.

If the updated measurement data N is different from the latest measurement data in the state change history 110, the state change recording unit 12 adds the updated measurement data N to the state change history 110 to create an updated state change history 110. . The state change recording section 12 and the data collection section 10 do not need to be separate, and may be one section that performs both functions.

The health determination unit 13 determines the health of the management point 4 by comparing the updated status change history 110 and the normal operation pattern group 111. In other words, the health determination unit 13 determines the health of the equipment 3 by comparing changes in the updated measurement data N at the management point 4 and the normal operation pattern based on a certain determination rule. Further, the soundness determining unit 13 may have a function of comparing the result determined by the administrator 6 and the determination made by the soundness determining unit 13 and learning by itself.

The alarm management unit 14 notifies the administrator 6 of an alarm when the health determination unit 13 determines that the updated state change history 110 does not correspond to the normal operation pattern group 111. The warning is, for example, an email, and is sent via the network 5 to a device such as a mobile phone carried by the administrator 6, so that the administrator 6 can be promptly notified. The manager 6 who received the alarm can determine whether there is an abnormal state by actually checking the state of the equipment 3 having the management point 4. By accessing the alarm history 112, the administrator 6 can check the presence or absence of past alarms and their contents, so the administrator 6 can also use the alarm history 112 as reference data when determining the health of the management point 4. good.

If the administrator 6 who has been notified of the alarm determines that the state is normal, the normal operation pattern creation unit 15 replaces the updated status change history 110 with at least a part of the normal operation pattern group 111 to create a new one. A new normal operation pattern group 111 is created by creating a normal operation pattern group 111, or by adding at least a part of the updated state change history 110 to the normal operation pattern group. Even if the equipment 3 is of the same type, the environment in which each equipment 3 is placed is different, so it is preferable that the health determination of each management point 4 can also be customized. Since the equipment monitoring device 1 is equipped with the normal operation pattern creation unit 15 and can create a new normal operation pattern group 111 in accordance with each management point 4, it is possible to appropriately determine the health of each management point 4. .

FIG. 2 is a hardware configuration diagram of a computer that constitutes the equipment monitoring device 1 in an example of the embodiment. The computer constituting the equipment monitoring device 1 can be realized with a general-purpose hardware configuration. That is, as shown in FIG. 2, the computer includes a CPU 21, a ROM 22, a RAM 23, an HDD controller 25 to which a hard disk drive (hereinafter referred to as HDD) 24 is connected, a mouse 26 and a keyboard 27 provided as input means, and a display device. It is constructed by connecting an input/output controller 29 to which a display 28 provided as a computer is connected, and a network controller 30 provided as a communication means to an internal bus 31.

The information storage unit 11 can be configured by one or more HDDs 24. In addition, in preparation for an emergency, it may further include a backup HDD 24. The data collection section 10, the state change recording section 12, the health determination section 13, the alarm management section 14, and the normal operation pattern creation section 15 can be configured by a CPU 21, a ROM 22, and a RAM 23. The administrator 6 accesses the status change history 110, normal operation pattern group 111, or alarm history 112 stored in the HDD 24 using the mouse 26, keyboard 27, and display 28, or via the network controller 30. You can.

Next, the state change history 110 stored in the information storage unit 11 will be described with reference to FIG. 3. The state change history 110 is a record of changes in measurement data collected by the data collection unit 10 at the management points 4 before updating. Five management points 4 are illustrated: lighting A setting, lighting A status, supply voltage A, power consumption A, and air conditioner A setting temperature. The four management points 4 of lighting A setting, lighting A status, supply voltage A, and air conditioner A setting temperature automatically send the occurrence time and updated measurement data N to the data collection unit 10 when there is a change in measurement data. You can also send it. The power consumption A management point 4 may transmit updated measurement data N to the data collection unit 10 every hour.

In FIG. 3, for each management point 4, occurrence times and measurement data at the occurrence times are arranged from top to bottom in order of measurement time. Although the number of measurement data for each management point 4 is illustrated as three, it is of course possible to use a number other than three, and the number of measurement data for each management point 4 may be different. In order to determine the health of the management point 4, the health determination unit 13 calculates the state change frequency of the measurement data, that is, the number of changes in the measurement data per hour, so that changes over an hour can be recorded. It is preferable to set the number of measurement data of the management point 4. The measurement data of each management point 4 in the state change history 110 may be automatically deleted after a certain period of time, or may be deleted as appropriate at the discretion of the administrator 6.

Next, the normal operation pattern group 111 stored in the information storage unit 11 will be explained with reference to FIG. The normal operation pattern group 111 is a set of normal operation patterns that should be determined to indicate that the management point 4 is healthy with respect to changes in the updated measurement data N. The normal operation patterns of changes in measurement data that can be determined to be that there is no abnormality or sign of abnormality are Lighting A setting pattern 1, Lighting A setting pattern 2, Lighting A state pattern 1, Supply voltage A pattern 1, and Power consumption. Six examples are shown: A/Pattern 1 and air conditioner A set temperature/Pattern 1. Regarding lighting A settings, two patterns, lighting A settings pattern 1 and lighting A settings pattern 2, are stored as normal operation patterns. Therefore, the health determination unit 13 determines whether the updated status change history 110 corresponds to either lighting A setting/pattern 1 or lighting A setting/pattern 2.

In FIG. 4, occurrence times and measurement data at the occurrence times are arranged in order of measurement time from top to bottom for each normal operation pattern. Although the number of measurement data for each normal operation pattern is exemplified as three, the number may be other than three, and the number of measurement data may be different for each normal operation pattern. Since the health determination unit 13 determines the health of the management point 4 by comparing the state change history 110 and the normal operation pattern group 111, the measurement data of each normal operation pattern included in the normal operation pattern group 111 is The amount can be such that it can be compared with the state change history 110.

The normal operation pattern includes an interval that indicates the time interval at which the measured data changes, a numerical fluctuation rate and state change frequency that indicate the state of change in the measured data before updating, and an occurrence time that extracts the time zone of the measurement time. Includes the band as data.

The numerical fluctuation rate is a numerical value calculated as the magnitude of fluctuation in measured data, and is calculated when the measured data is a continuous value such as supply voltage, power consumption, air conditioner set temperature, etc. For example, the numerical fluctuation rate in the second and third columns from the top of supply voltage A pattern 1 is 1.7% based on the difference from the measurement data in the first and second columns from the top, respectively. It is calculated as -1.6%.

The state change frequency indicates the number of changes in measurement data per hour, and is calculated whether the measurement data is a continuous value or a discrete value. For example, in supply voltage A and pattern 1, the measurement data fluctuates three times in one hour from midnight to 1 am.

The occurrence time period indicates the time period when the measurement data changes in units of one hour. For example, the measurement data for supply voltage A and pattern 1 are all collected during one hour from 0:00 to 1:00.

FIG. 5 shows an example of a normal operation pattern group 111 including measurement data of other management points 4. Normal operation patterns include Lighting A setting pattern 1, Lighting A setting pattern 2, Lighting A status pattern 1, Supply voltage A pattern 1, Power consumption A pattern 1, and Air conditioning A setting temperature pattern 1. One example is shown below.

The normal operation pattern group 111 shown in FIG. 5 is the same as the normal operation pattern group 111 shown in FIG. The latest history of entering and leaving the room has been added. In other words, the normal operation pattern of one management point 4 included in the normal operation pattern group 111 includes measurement data of one or more other management points 4 or data related to the operation of the equipment 3. Can be done. By checking the entry/exit history of door A near lighting A, it may be possible to detect abnormalities or signs of abnormalities that could not be detected conventionally. Further, for example, data acquired by a device other than the equipment monitoring device 1, such as measurement data regarding the temperature and humidity outside the facility 2, can also be included in the normal operation pattern group 111.

Next, the equipment monitoring method will be explained in detail with reference to FIG. FIG. 6 is a flowchart showing the process when the data collection unit 10 collects updated measurement data N.

When the equipment monitoring device 1 collects the updated measurement data N at the management point 4 (step S101), the updated measurement data N is the latest data in the state change history 110 that records changes in the measurement data at the management point 4 before updating. It is determined whether the measured data is different from the measured data (step S102).

If the updated measurement data N is the same as the state change history 110 (NO in step S102), the equipment monitoring device 1 ends the process and waits until new updated measurement data N is collected. On the other hand, if the updated measurement data N is different from the state change history 110 (YES in step S102), the equipment monitoring device 1 proceeds to step S103.

In step S103, the equipment monitoring device 1 adds the updated measurement data N to the state change history 110 to create an updated state change history 110. After that, the equipment monitoring device 1 determines whether the updated state change history 110 corresponds to the normal operation pattern group 111, which is a set of normal operation patterns (step S104).

If the updated measurement data N takes a discrete value and the interval or state change frequency deviates by a certain percentage or more compared to the normal operation pattern, the updated state change history 110 does not correspond to the normal operation pattern group 111. can be determined. For example, in an example of the normal operation pattern group 111 in FIG. 4, a normal operation pattern in which the frequency of state changes in one hour from 17:00 to 18:00 is 2 times/hour is recorded as illumination A state/pattern 1. Therefore, if the frequency of state changes in one hour from 17:00 to 18:00 exceeds 3 times/hour, chattering due to contact failure is suspected. It can be determined that the pattern does not correspond to the pattern group 111.

In addition, if the updated measurement data N takes a continuous value and the absolute value of the numerical fluctuation rate or the state change frequency deviates by a certain percentage or more compared to the normal operation pattern, the updated state change history 110 will be based on the normal operation pattern. It can be determined that it does not correspond to group 111. For example, in Figure 4, for power consumption A/pattern 1, power consumption A for one hour from 9:00 to 10:00 is normal, increasing by 24.4% compared to power consumption A for one hour from 8:00 to 9:00. Operational patterns are recorded. Regarding power consumption A, a fluctuation of about ±24.4% is within the normal numerical range, but if it exceeds this range, a malfunction of the meter is suspected. It can be determined that the pattern does not correspond to the normal operation pattern group 111.

If it is determined that the updated status change history 110 corresponds to the normal operation pattern group 111 (YES in step S104), the equipment monitoring device 1 ends the process and newly collects updated measurement data N. Wait until. On the other hand, if it is determined that the updated state change history 110 does not correspond to the normal operation pattern group 111 (NO in step S104 and step S105), the equipment monitoring device 1 notifies the administrator 6 of an alarm ( Step S106).

The manager 6 who has been notified of the warning checks the status of the management point 4 (step S107). At this time, the administrator 6 may not only directly observe the management points 4 but also check the status change history 110, normal operation pattern group 111, and alarm history 112 stored in the information storage unit 11. If the administrator 6 determines that the management point 4 is in an abnormal state (YES in step S108), the administrator 6 takes measures to resolve the abnormal state, that is, repairs the equipment 3 having the management point 4. (Step S111).

On the other hand, if the administrator 6 determines that the management point 4 is in a normal state (NO in step S108), the administrator 6 further sets the status change history determined to be in a normal state as a normal operation pattern. It is determined whether to register in group 111 (step S109). As a result, when the range of judgments as normal is expanded compared to before, by adding a new normal operation pattern, the health judgment unit 13 can judge the health to be healthy in more cases, which reduces the burden on the administrator 6. be able to. Furthermore, if the administrator 6 determines that the state is normal but is not sure, he or she can temporarily postpone registering it as a normal operation pattern. If it is not registered as a normal operation pattern in the normal operation pattern group 111 (NO in step S109), the equipment monitoring device 1 ends the process and waits until new updated measurement data N is collected. Note that step S107, step S108, step S109, and step S111 are processes performed by the administrator 6. In FIG. 6, the processing performed by the administrator 6 is indicated by a broken line, and is distinguished from other processing performed by the facility monitoring device 1.

When receiving an instruction to register as a normal operation pattern in the normal operation pattern group 111 (YES in step S109), the equipment monitoring device 1 stores at least a portion of the normal operation pattern group 111 in the state change history 110 after updating. Create a new normal operation pattern group 111 by replacing at least a part of the updated status change history 110, or create a new normal operation pattern group 111 by adding at least a part of the updated state change history 110 to the normal operation pattern group 111. Create (step S110). After that, the equipment monitoring device 1 ends the process and waits until new updated measurement data N is collected.

Note that the administrator 6 may be a computer instead of a human.

As described above, according to the equipment monitoring device 1 and the equipment monitoring method of this embodiment, abnormalities or signs of abnormalities that could not be detected conventionally can be detected.

1 equipment monitoring device, 2 facility, 3 equipment, 4 management point, 5 network, 6 administrator, 10 data collection unit, 11 information storage unit, 110 status change history, 111 normal operation pattern group, 112 alarm history, 12 status change Recording unit, 13 Soundness determination unit, 14 Alarm management unit, 15 Normal operation pattern creation unit, 21 CPU, 22 ROM, 23 RAM, 24 Hard disk drive (HDD), 25 HDD controller, 26 Mouse, 27 Keyboard, 28 Display, 29 input/output controller, 30 network controller, 31 internal bus, N updated measurement data.

Claims (4)

a data collection unit that collects updated measurement data updated at the management point;
an information storage unit that stores a state change history that records changes in measurement data before updating at the management point, and a normal operation pattern group that is a set of normal operation patterns for determining the health of the management point;
a state change recording unit that adds the updated measurement data to the state change history to create an updated state change history when the updated measurement data is different from the state change history;
a health determination unit that compares the updated status change history and the normal operation pattern group to determine the health of the management point;
an alarm management unit that notifies an administrator of an alarm when the health determination unit determines that the state change history after the update does not correspond to the normal operation pattern group ;
If the administrator determines that the management point is in a normal state, a new normal operation pattern group is created by replacing at least a portion of the normal operation pattern group with at least a portion of the updated status change history. or a normal operation pattern creation unit that creates a new normal operation pattern group by adding at least a part of the updated state change history to the normal operation pattern . The equipment monitoring device according to claim 1, wherein the information storage unit further stores an alarm history in which the alarms are recorded. The normal operation pattern of the management point included in the normal operation pattern group includes measurement data of one or more other management points before updating or data related to the operation of equipment having the management point. , The equipment monitoring device according to claim 1 or 2 . a data collection step of collecting updated measurement data at the management point;
If the updated measurement data is different from a state change history that records changes in measurement data before the update at the management point, add the updated measurement data to the state change history to create a state change history after the update. a state change history creation step;
a health determination step of determining the health of the management point by comparing the updated state change history with a group of normal operation patterns for determining the health of the management point;
If it is determined in the health determination step that the updated state change history does not correspond to the normal operation pattern group, an alarm notification step of notifying an administrator of an alarm ;
If the administrator determines that the management point is in a normal state and further determines that the status change history after the update should be registered as a normal operation pattern, at least a part of the normal operation pattern group is updated as described above. A new normal operation pattern group is created by replacing at least a part of the later status change history, or a new normal operation pattern is created by adding at least a part of the updated status change history to the normal operation pattern. A method for monitoring equipment, including a step of creating a normal operation pattern to create a group .

Images