Machine monitoring and diagnosis device, its monitoring and diagnosis method, and machine monitoring and diagnosis system

Abstract

To display information of correlation between newly generated parameters on an output device when an event which is difficult to model in advance occurs.SOLUTION: A monitoring diagnostic device of an apparatus comprises: a measurement database during normal time in which measurement data when the apparatus is normal are stored; a measurement data input device in which current measurement data of the apparatus are input; a parameter correlation calculation unit to which the measurement data during normal time and the current measurement data are inputted, and which calculates strength of correlation between parameters during normal time and at current time, and identifies an abnormal parameter by calculating a variation of correlation between parameters during normal time and at current time, and detecting the existence of abnormality; a correlation change direction determination unit which distinguishes an expected event at which the correlation between the parameters is known to break down from an unexpected event when there is the correlation between the parameters during normal time and an abnormality has occurred; and a first result output device which outputs and displays a variation of the correlation between the parameters when the correlation change direction determination unit determines as the unexpected event.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a monitoring and diagnostic device for equipment such as plant equipment, a monitoring and diagnostic method thereof, and a monitoring and diagnostic system for equipment.

Background Art

[0002] In recent years, in chemical plants and power generation plants, for the purpose of stable operation and rationalization of maintenance work, the introduction of condition monitoring technology during plant equipment operation has been promoted.

[0003] When an abnormality occurs in plant equipment, by detecting the abnormality of the plant equipment at an early stage and taking countermeasures, it is possible to expect continuous operation and shortening of maintenance work time.

[0004] As a method for detecting an abnormality in plant equipment, there is a method of determining the presence or absence of an abnormality by comparing plant measurement values such as temperature, pressure, vibration, and rotational speed with past normal measurement data. For example, there is a correlation between the rotational speed and vibration of a rotating machine. If shaft eccentricity occurs, the vibration increases even at the same rotational speed. Therefore, an abnormality can be determined based on a change in the correlation.

[0005] As such a method for determining the presence or absence of an abnormality, a method of approximating the relationship between parameters with a regression curve and using the distance from the curve as the degree of abnormality, or a method of calculating the Mahalanobis distance and using it as the degree of abnormality is generally known.

[0006] As an example thereof, there is an abnormality determination device for a compressor of a gas turbine described in Patent Document 1.

[0007] In this Patent Document 1, in order to determine an abnormality of a target device in view of the correlation of a plurality of parameters, a group acquisition unit that acquires values of at least one parameter group including two or more parameters having a correlation with each other related to the target device, a distance specifying unit that specifies a distance between a reference line representing the correlation between the parameters constituting the parameter group and the acquired values of the parameter group, and an output unit that outputs an alarm when the specified distance exceeds a predetermined range are provided in an abnormality determination device.

[0008] Also, when an abnormality is detected, it is effective to present which parameter has the abnormality, but in order to perform countermeasures and maintenance work, it is desirable to present the cause of the abnormality.

[0009] In the above example, it is not only the result that there is an abnormality in the vibration, but also to show that it is caused by the eccentricity of the shaft.

[0010] As an example of such a cause estimation method, there is a device condition monitoring system in a power generation plant or the like described in Patent Document 2.

[0011] This Patent Document 2 includes a normal measurement value database storing normal measurement values of a plant when the plant equipment is normal, a device deterioration model database storing a device deterioration model in which the relationship between parameters affected during device failure is modeled, a device failure probability database storing the failure probability of the device, and a device failure record database storing the failure records of the device. A physical model setting unit that sets a physical model by setting a probability calculated from the device failure probability database and the device failure record database in the device deterioration model, an abnormal parameter estimation unit that acquires measurement values of the plant and estimates abnormal parameters by comparing them with the normal measurement values stored in the normal measurement value database, and a failed device estimation unit that estimates a failed device from the abnormal parameters using the physical model are provided in a device condition monitoring system.

Prior Art Documents

Patent Document

[0012]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0013] When using the abnormality determination device described in Patent Document 1 above, abnormal parameters can be identified by comparing plant measurement values with normal measurement values. When using the equipment status monitoring system described in Patent Document 2, the cause of the abnormality can be estimated from the abnormal parameters. However, it is applicable only to those in which there is a correlation between parameters during normal operation and the correlation between those parameters breaks down when an abnormality occurs.

[0014] As an example where it is difficult to handle with the technologies described in Patent Documents 1 and 2 like this, consider the event of steam leaking from a pipe near a rotating machine.

[0015] For example, during normal operation, there is no direct correlation between the bearing temperature of the rotating machine and the pressure of the pipe near the rotating machine. However, if a leak occurs in the pipe and steam comes into contact with the rotating machine, a correlation may occur where the pipe pressure decreases and at the same time the bearing temperature rises.

[0016] Thus, between parameters that have no correlation during normal operation, a reference line representing the correlation cannot be defined, so the technology of Patent Document 1 cannot be used. Also, such events occur infrequently and it is difficult to predict in advance whether the influence on the parameters will appear, so it is difficult to prepare the physical model of Patent Document 2.

[0017] An event in which there is a correlation between parameters under normal conditions and the correlation between these parameters breaks down when an abnormality occurs is hereinafter referred to as an assumed event. On the other hand, an event in which a correlation occurs between parameters that have no correlation under normal conditions when an abnormality occurs is hereinafter referred to as an unexpected event.

[0018] The technologies described in Patent Documents 1 and 2 mentioned above are both targeted at assumed events, and it is difficult to identify the cause of an abnormality for unexpected events. However, even in the case of unexpected events, information on which parameters a correlation has occurred between can be utilized for human cause investigation.

[0019] In the example described above, if it is found that a new correlation has occurred between the pipeline pressure and the bearing temperature of the rotating machine, the possibility of a leak from the pipeline can be suspected. However, since it is difficult to utilize without specialized knowledge such as that of engineers, it is necessary to devise the information providing destination and the providing method.

[0020] Patent Documents 1 and 2 do not describe anything about how to present information on the correlation between parameters in such unexpected events.

[0021] The present invention has been made in view of the above points, and its object is to provide a monitoring and diagnostic device for equipment, a monitoring and diagnostic method for the equipment, and a monitoring and diagnostic system for equipment that can present information on the correlation between newly generated parameters to an output device when an event that is difficult to model in advance occurs.

Means for Solving the Problem

[0022] The monitoring and diagnostic device for equipment of the present invention is a monitoring and diagnostic device for equipment that detects an abnormality of the equipment by a change in parameter correlation, and A normal measurement database storing measurement data when the device is normal, a measurement data input device into which current measurement data of the device is input, the measurement data at normal times from the normal measurement database and the current measurement data from the measurement data input device are input, the strength of the correlation between the parameters at normal times and currently is calculated for these, the amount of change in the correlation between the parameters at normal times and currently is calculated to detect the presence or absence of an abnormality and identify abnormal parameters, a parameter correlation calculation unit, when a correlation exists between the parameters at normal times and an abnormality occurs, an assumed event in which it is known that the correlation between these parameters breaks down and an unexpected event in which a correlation occurs between the parameters that have no correlation at normal times when an abnormality occurs are distinguished by a correlation change direction determination unit, and when the correlation change direction determination unit determines that it is an unexpected event, a first result output device that outputs and displays the amount of change in the correlation between the parameters. It is characterized by comprising the above.

[0023] Also, a method for monitoring and diagnosing a device according to the present invention is a method for monitoring and diagnosing a device that detects an abnormality of the device by a change in parameter correlation in order to achieve the above object, The step of storing the measurement data when the device is normal in the normal-time measurement database; the step of inputting the current measurement data of the device into the measurement data input device; the normal-time measurement data from the normal-time measurement database and the current measurement data from the measurement data input device are input into the parameter correlation calculation unit, and the parameter correlation calculation unit calculates the strength of the correlation between each parameter at normal time and the current time for the normal-time measurement data and the current measurement data, calculates the change amount of the correlation between each parameter at normal time and the current time to detect the presence or absence of an abnormality; when there is a correlation relationship between each parameter at normal time and an abnormality occurs, the assumed event that the correlation relationship between these parameters breaks down and the assumed unexpected event that a correlation relationship occurs between each parameter that has no correlation relationship at normal time when an abnormality occurs are distinguished by the correlation change direction determination unit; when it is determined by the correlation change direction determination unit as the assumed unexpected event, the change amount of the correlation between each parameter is output and displayed on the first result output device.

[0024] Also, the device monitoring and diagnosis system of the present invention, in order to achieve the above object, the measured value of the device measured by a sensor installed at a predetermined location of the device is transmitted to the device monitoring and diagnosis device via a network, and the cause of the abnormality of the device detected by the change of the parameter correlation by the device monitoring and diagnosis device based on the transmitted measured value is transmitted to the owner of the device. The device monitoring and diagnosis system is characterized in that the device monitoring and diagnosis device is the device monitoring and diagnosis device having the above configuration.

Effect of the Invention

[0025] According to the present invention, when an event that is difficult to model in advance occurs, it is possible to present information on the correlation relationship between newly generated parameters to an output device.

Brief Description of the Drawings

[0026]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

MODE FOR CARRYING OUT THE INVENTION

[0027] Hereinafter, a monitoring and diagnostic apparatus for equipment, a monitoring and diagnostic method therefor, and a monitoring and diagnostic system for equipment according to the present invention will be described based on the illustrated embodiments. In each embodiment, the same reference numerals are used for the same components.

Embodiment

[0028] Fig. 1 shows the overall configuration of a centrifugal compressor 11 to which the monitoring and diagnostic apparatus for equipment of the present invention is applied.

[0029] As shown in Fig. 1, the centrifugal compressor 11 is generally composed of a casing 12 that is usually formed in a cylindrical shape and serves as a stationary part, a rotor 15 disposed within the casing 12, and a journal bearing 13 and a thrust bearing 14 that rotatably support the rotor 15. A plurality of impellers 16 are provided on the above-described rotor 15, and the fluid flowing in the flow path is compressed by the rotation of the impellers 16.

[0030] For the centrifugal compressor 11 having the above configuration, sensors for measuring the temperature, pressure, vibration, rotational speed, etc. of the centrifugal compressor 11 are appropriately installed at locations suitable for each measurement.

[0031] The sensor 17 shown in Fig. 1 is installed on the rotor 15 to measure shaft vibration as an example. The measured value (shaft vibration) of the centrifugal compressor 11 measured by this sensor 17 is transmitted to a monitoring system 20 via networks such as a DCS (Distributed Control System) 18 and a LAN 19. When monitoring the state of the centrifugal compressor 11 from the monitoring system 20 and estimating the cause of an abnormality when there is an abnormality, it is transmitted to a monitoring and diagnostic apparatus 10 for equipment that estimates the abnormality based on the change in parameter correlation. The abnormal parameters and the cause of the abnormality of the centrifugal compressor 11 estimated by the monitoring and diagnostic apparatus 10 based on the measured values measured by the sensor 17 are transmitted to the owner (contract user) 21 of the centrifugal compressor 11, thus configuring a monitoring and diagnostic system for equipment.

[0032] Next, the above-described monitoring and diagnostic apparatus 10 for equipment will be described with reference to Fig. 2. Fig. 2 is a diagram showing Embodiment 1 of the monitoring and diagnostic apparatus for equipment of the present invention.

[0033] As shown in Fig. 2, the monitoring and diagnosis device 10A of the equipment in this embodiment includes a normal measurement database 2 storing measurement data when the centrifugal compressor 11, which is the equipment, is normal, a measurement data input device 1 into which the current measurement data of the centrifugal compressor 11 is input, and a parameter correlation calculation unit 4 into which the normal measurement data from the normal measurement database 2 and the current measurement data from the measurement data input device 1 are input, and which calculates the strength of the correlation between each parameter at normal time and currently, calculates the change amount of the correlation between each parameter at normal time and currently, detects the presence or absence of an abnormality, and identifies abnormal parameters; a correlation change direction determination unit 5 that distinguishes between an assumed event in which the correlation between each parameter collapses when an abnormality occurs although there is a correlation between each parameter at normal time, and an unexpected event in which a correlation occurs between parameters that have no correlation at normal time when an abnormality occurs; a first result output device 7 that outputs and displays the change amount of the correlation between each parameter when it is determined as an unexpected event by the correlation change direction determination unit 5; an abnormal cause database 3 storing the relationship between the failure mode of the centrifugal compressor 11 and abnormal parameters; a cause estimation unit 6 that estimates the abnormal cause from the abnormal parameters stored in the abnormal cause database 3 when it is determined as an assumed event by the correlation change direction determination unit 5; and a second result output device 8 that outputs and displays the abnormal cause estimated by the cause estimation unit 6.

[0034] More specifically, in the monitoring and diagnosis device 10A of the equipment in this embodiment shown in Fig. 2, current measurement data is input into the measurement data input device 1. Signals such as temperature, pressure, vibration, and rotational speed measured by measurement devices (such as sensors) installed in the plant or equipment are input. Fig. 3 shows an example of the measurement data at each time in 2010.

[0035] In addition, the normal measurement database 2 stores measurement data when the equipment is normal. The normal measurement database 2 is the measurement data when the equipment is normal at each time in 2010, similar to the measurement data in Fig. 3.

[0036] An example of graphing the measurement data (vibration and rotational speed) of the above assumed events (events where there is a correlation between parameters during normal operation and the correlation between those parameters breaks when an abnormality occurs) is shown in FIG. 4.

[0037] As shown in FIG. 4, during normal operation, there is a correlation between vibration and rotational speed. However, when an abnormality such as shaft eccentricity occurs, the vibration increases with respect to the rotational speed, and the correlation between vibration and rotational speed breaks down.

[0038] On the other hand, an example of graphing the measurement data (temperature and pressure) of an unexpected event (an event where a correlation occurs between parameters that have no correlation during normal operation when an abnormality occurs) is shown in FIG. 5.

[0039] As shown in FIG. 5, during normal operation, there is no correlation between temperature and pressure. However, when an abnormality such as steam leakage occurs from a pipe, the values of temperature and pressure change simultaneously, so a correlation occurs.

[0040] In addition, the abnormal cause database 3 stores the relationship between the failure mode of the equipment and the abnormal parameters. An example of the abnormal cause database 3 is shown in FIG. 6.

[0041] The abnormal cause database 3 shown in FIG. 6 is prepared in advance for assumed events. In the example shown in FIG. 6, it means that when shaft eccentricity occurs in a rotating machine, there may be an abnormality in vibration, and when bearing wear occurs in a rotating machine, there may be abnormalities in vibration and temperature. Similarly, it means that when leakage occurs in a valve, there may be an abnormality in pressure.

[0042] In addition, the parameter correlation calculation unit 4 receives the current measurement data from the measurement data input device 1 and the normal measurement data from the normal measurement database 2, calculates the strength of the correlation between each parameter, calculates the change amount of the correlation between normal times and the current time to detect the presence or absence of an abnormality, and identifies the abnormal parameters.

[0043] For example, the strength of correlation can be calculated by computing the absolute value of the correlation coefficient between each parameter, and the amount of change can be determined by calculating the difference therebetween.

[0044] Taking the measurement data from 00:00:00 to 00:23:00 on January 1, 2010 in the measurement data of FIG. 3 as the normal measurement data, FIG. 7 shows an example of the result of calculating the absolute value of the correlation coefficient between each parameter.

[0045] As is apparent from FIG. 7, it can be seen that during normal operation, there is a large correlation between the rotational speed and the temperature, and between the rotational speed and the vibration, and the pressure has a small correlation with the other parameters (temperature, vibration, rotational speed).

[0046] FIG. 8 shows an example of the result of calculating the absolute value of the correlation coefficient of the current measurement data in the same manner.

[0047] As is apparent from FIG. 8, it can be seen that in the correlation relationship of the current measurement data, there is a large correlation between the rotational speed and the temperature, and the pressure has a small correlation with the other parameters (temperature, vibration, rotational speed).

[0048] From these results, the change in the correlation relationship between the normal time and the current time can be calculated as shown in FIG. 9. That is, for anomaly detection, for example, when the change is 0.5 or more, the parameter is detected as an anomaly. In this example, as can be seen from FIG. 9, the vibration or the rotational speed has changed by 0.5 or more, and the abnormal parameter can be identified by detecting that there is an anomaly in the vibration or the rotational speed.

[0049] In addition, the correlation change direction determination unit 5 receives the amount of change in the correlation with the abnormal parameter from the parameter correlation calculation unit 4 and determines the direction (increase or decrease) of the correlation change. That is, in the correlation change direction determination unit 5, when the correlation decreases by a certain amount or more, it is determined as an assumed event, and when the correlation increases by a certain amount or more, it is determined as an unexpected event.

[0050] Further, in the cause estimation unit 6, when an event is determined as an assumed event by the correlation change direction determination unit 5, the relationship between the failure mode of the device and the abnormal parameter is input from the abnormal cause database 3, and the abnormal cause is estimated from the abnormal parameter. In the above example, when there is an abnormality in vibration or rotational speed, referring to FIG. 6, when vibration occurs, it is estimated that the cause is shaft eccentricity or bearing wear.

[0051] Also, the second result output device 8 is a display installed in the plant, and the abnormal parameter and the abnormal cause are input from the cause estimation unit 6, and the diagnosis result is output and displayed.

[0052] FIG. 10 is an example of the output display screen of the second result output device 8, where shaft vibration and efficiency are displayed as abnormal parameters, and flow path fouling is displayed as the abnormal cause at that time.

[0053] Since the main targets for outputting and displaying the diagnosis result on the second result output device 8 are the operator and the maintenance staff, it is very effective to present the abnormal cause as the information necessary for dealing with the occurrence of an abnormality.

[0054] On the other hand, when an event is determined as an unexpected event by the correlation change direction determination unit 5, the diagnosis result is output and displayed on the first result output device 7.

[0055] This first result output device 7 is a display installed in the office of the equipment manufacturer outside the plant, and the abnormal parameter and the amount of change in the correlation are input from the correlation change direction determination unit 5, and the diagnosis result is output and displayed.

[0056] FIG. 11 is an example of the output screen of the first result output device 7, where temperature and pressure are displayed as abnormal parameters, and the amount of change in the correlation between the temperature and the pressure at that time is displayed.

[0057] Since the main target for outputting and displaying the diagnosis result on the first result output device 7 is the manufacturer's engineer, even in the case of an unexpected event where the cause cannot be identified, it is possible to examine the abnormal cause and the necessity of countermeasures from the change in the correlation with the abnormal parameter.

[0058] Note that the parameter correlation calculation unit 4, the correlation change direction determination unit 5, and the cause estimation unit 6 may be implemented as a computer program, or the normal measurement database 2 and the abnormal cause database 3 may be included in the computer.

[0059] FIG. 12 is a flowchart for explaining the processing (monitoring and diagnosis method) by the device monitoring and diagnosis apparatus 10A of the present embodiment.

[0060] In FIG. 12, in step S1, the measurement data at normal times is input from the normal measurement database 2 to the parameter correlation calculation unit 4, and in step S2, the current measurement data is input from the measurement data input device 1 to the parameter correlation calculation unit 4.

[0061] In step S3, the parameter correlation calculation unit 4 calculates the correlation between the parameters for the normal measurement data and the current measurement data. In step S4, if there is no change in the correlation between the parameters, the process returns to step S2 to continue the measurement.

[0062] In step S5, the correlation change direction determination unit 5 calculates the change amount of the correlation between the normal measurement data and the current measurement data. If the correlation decreases, it is determined that there was a correlation relationship between the parameters at normal times and that the correlation relationship between those parameters breaks down when an abnormality occurs, and the process proceeds to step S6. If the correlation increases, it is determined that an unexpected event occurs in which a correlation relationship occurs between parameters that had no correlation relationship at normal times when an abnormality occurs, and the process proceeds to step S8.

[0063] In step S6, the cause estimation unit 6 estimates the abnormal cause from the abnormal parameters. In step S7, the abnormal parameters and the abnormal cause are output and displayed on the second result output device 8 as the diagnosis result to the second output destination. In step S8, the abnormal parameters and the change amount of the correlation are output and displayed on the first result output device 7 as the diagnosis result to the first output destination.

[0064] According to such an embodiment of the present invention, not only can the cause of an abnormality in an assumed event be estimated, but also information on the correlation between newly generated parameters can be presented to the output device even when an event that is difficult to model in advance occurs.

Embodiment

[0065] FIG. 13 is a configuration diagram showing Embodiment 2 of the device monitoring and diagnosing apparatus of the present invention.

[0066] The device monitoring and diagnosing apparatus 10B of the present embodiment shown in FIG. 13 is configured such that the first result output device 7 can present both the information of the second result output device 8 and the information of the first result output device 7.

[0067] That is, the abnormal parameter and the cause of the abnormality estimated by the cause estimation unit 6 are output and displayed on the second result output device 8 and also output and displayed on the first result output device 7.

[0068] According to such an embodiment of the present invention, of course, the same effects as those of Embodiment 1 can be obtained. Not only is the change amount of the correlation between the abnormal parameter and the diagnosis result output and displayed to the first output destination, but also the abnormal parameter and the cause of the abnormality can be output and displayed, so that the amount of information obtained increases.

Embodiment

[0069] Although not particularly illustrated, Embodiment 3 of the device monitoring and diagnosing apparatus of the present invention is a device monitoring and diagnosing apparatus in which the first result output device 7 and the second result output device 8 are implemented as the same device, and the information presentation function as the first result output device 7 and the information presentation function as the second result output device 8 are switched by the user and output and displayed.

[0070] For example, as the device monitoring and diagnosing apparatus of the present embodiment, information as the second result output device 8 is presented to drivers and maintenance personnel, and information as the first result output device 7 is presented to upper-level drivers, upper-level maintenance personnel, engineers, and the like.

[0071] According to such an embodiment of the present invention, the same effects as those of Embodiment 1 can be obtained. Moreover, since only one result output device is required, the number of components can be reduced.

Embodiment

[0072] Embodiment 4 of the monitoring and diagnostic device for equipment of the present invention has the same configuration and processing content (monitoring and diagnostic method) as Embodiment 1 described above, but is a monitoring and diagnostic device for equipment in which the criterion for determining abnormality in the parameter correlation calculation unit 4 is set lower when the correlation increases.

[0073] That is, the criterion for detecting the presence or absence of abnormality in the parameter correlation calculation unit 4 is distinguished by the correlation change direction determination unit 5, and information is presented to the first result output device 7 at a timing earlier than that of the second result output device 8.

[0074] For example, when the correlation increases by 0.3 or more or decreases by 0.5 or more, the parameter is detected as abnormal. As a result, when an unexpected event occurs, it can be detected earlier. Therefore, for drivers and maintenance personnel, unnecessary responses can be reduced by presenting information after the accuracy of the abnormality increases. For engineers, even if the accuracy of the abnormality is low, the signs can be presented earlier.

[0075] It is also within the scope of the present invention to regard measurement data for which abnormal data has not been extracted or past measurement data that has not been processed for abnormal extraction as the "normal measurement data" in each of the above-described embodiments.

[0076] In addition, the present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments have been described in detail for the purpose of easily explaining the present invention, and are not necessarily limited to those having all the configurations described. Also, each of the above configurations, functions, etc. may be realized by designing part or all of them, for example, by an integrated circuit. Further, each of the above configurations, functions, etc. may be realized by software by a processor interpreting and executing a program for realizing each function.

Explanation of Reference Numerals

[0077] 1... Measurement data input device, 2... Normal measurement database, 3... Abnormal cause database, 4... Parameter correlation calculation unit, 5... Correlation change direction determination unit, 6... Cause estimation unit, 7... First result output device, 8... Second result output device, 10, 10A, 10B... Equipment monitoring and diagnosis device, 11... Centrifugal compressor, 12... Casing, 13... Journal bearing, 14... Thrust bearing, 15... Rotor, 16... Impeller, 17... Sensor, 18... DCS (Distributed Control System), 19... LAN, 20... Monitoring system, 21... Owner of the centrifugal compressor.

Claims

1. A monitoring and diagnostic device for a machine that detects abnormalities in the machine based on changes in parameter correlations, a normal measurement database storing measurement data when the machine is normal, a measurement data input device into which current measurement data of the machine is input, wherein the normal measurement data from the normal measurement database and the current measurement data from the measurement data input device are input, and the strength of the correlation between each parameter at normal times and currently is calculated, and the amount of change in the correlation between each parameter at normal times and currently is calculated to detect the presence or absence of an abnormality and identify abnormal parameters, a parameter correlation calculation unit; a correlation change direction determination unit that distinguishes between an assumed event in which the correlation between each parameter collapses when an abnormality occurs although there is a correlation between each parameter at normal times, and an unexpected event in which a correlation occurs between each parameter that has no correlation at normal times when an abnormality occurs; a first result output device that outputs and displays the amount of change in the correlation between each parameter when the correlation change direction determination unit determines the unexpected event, characterized in that it comprises a monitoring and diagnostic device for a machine.

2. The monitoring and diagnostic device for a machine according to claim 1, an abnormal cause database storing the relationship between the failure mode of the machine and abnormal parameters, a cause estimation unit that estimates the cause of the abnormality from the abnormal parameters stored in the abnormal cause database when the correlation change direction determination unit determines the assumed event, a second result output device that outputs and displays the cause of the abnormality estimated by the cause estimation unit, characterized in that it further comprises a monitoring and diagnostic device for a machine.

3. The monitoring and diagnostic device for a machine according to claim 2, wherein, in the cause estimation unit, when the correlation change direction determination unit determines the assumed event, the relationship between the failure mode of the machine and the abnormal parameters is input from the abnormal cause database, and the cause of the abnormality is estimated from the abnormal parameters, characterized in that it is a monitoring and diagnostic device for a machine.

4. The monitoring and diagnostic device for a machine according to claim 2, the assumed event determined by the correlation change direction determination unit is an event in which the current correlation has decreased by a certain amount or more between parameters that had a correlation of a certain amount or more at normal times, The monitoring and diagnostic device for equipment is characterized in that the assumed external event is an event when the current correlation increases by a certain amount or more between parameters that had little correlation during normal operation.

5. A monitoring and diagnostic device for equipment according to any one of claims 2 to 4, wherein the second result output device is installed inside the facility of the plant to be diagnosed, abnormal parameters and an abnormal cause are input from the cause estimation unit, and a diagnostic result is output and displayed; A monitoring and diagnostic device for equipment, wherein the first result output device is installed outside the facility of the plant to be diagnosed, abnormal parameters and the amount of change in correlation are input from the correlation change direction determination unit, and a diagnostic result is output and displayed.

6. A monitoring and diagnostic device for equipment according to any one of claims 2 to 5, wherein the first result output device is capable of outputting and displaying both the information of the second result output device and the information of the first result output device.

7. A monitoring and diagnostic device for equipment according to any one of claims 2 to 5, wherein the first result output device and the second result output device are the same device, and the information presentation function as the first result output device and the information presentation function as the second result output device are switched by the user and output and displayed.

8. A monitoring and diagnostic device for equipment according to any one of claims 2 to 7, wherein the criteria for detecting the presence or absence of abnormality in the parameter correlation calculation unit are distinguished by the correlation change direction determination unit, and information is presented to the first result output device at a timing earlier than that of the second result output device.

9. A monitoring and diagnostic method for equipment that detects equipment abnormalities based on changes in parameter correlation, including the steps of storing the measurement data when the equipment is normal in a normal-time measurement database; inputting the current measurement data of the equipment into a measurement data input device; inputting the normal-time measurement data from the normal-time measurement database and the current measurement data from the measurement data input device into a parameter correlation calculation unit, and in the parameter correlation calculation unit, calculating the strength of the correlation between each parameter at normal time and currently for the normal-time measurement data and the current measurement data, and calculating the amount of change in the correlation between each parameter at normal time and currently to detect the presence or absence of abnormality. When there is a correlation between the parameters under normal conditions and an abnormality occurs, it is known that the correlation between these parameters breaks down. The correlation change direction determination unit distinguishes between an assumed event and an unexpected event where a correlation occurs between parameters that have no correlation under normal conditions when an abnormality occurs. When the correlation change direction determination unit determines that it is an unexpected event, the amount of change in the correlation between the parameters is output and displayed on a first result output device. A monitoring and diagnostic method for a device, characterized by performing the above steps.

10. The monitoring and diagnostic method for a device according to claim 9, The step of storing the relationship between the failure mode of the device and the abnormal parameter in an abnormal cause database. When the correlation change direction determination unit determines that it is an assumed event, the cause estimation unit estimates the cause from the abnormal parameters stored in the abnormal cause database. The method further includes the step of outputting and displaying the abnormal cause estimated by the cause estimation unit on a second result output device. A monitoring and diagnostic method for a device, characterized by performing the above steps.

11. The monitoring and diagnostic method for a device according to claim 10, In the cause estimation unit, when the correlation change direction determination unit determines that it is an assumed event, the relationship between the failure mode of the device and the abnormal parameter is input from the abnormal cause database, and the cause is estimated from the abnormal parameter. A monitoring and diagnostic method for a device, characterized by the above.

12. The monitoring and diagnostic method for a device according to claim 10 or 11, The second result output device is installed inside the facility of the plant to be diagnosed. The abnormal parameter and the abnormal cause are input from the cause estimation unit, and the diagnostic result is output and displayed. The first result output device is installed outside the facility of the plant to be diagnosed. The abnormal parameter and the amount of change in the correlation are input from the correlation change direction determination unit, and the diagnostic result is output and displayed. A monitoring and diagnostic method for a device, characterized by the above.

13. The monitoring and diagnostic method for a device according to any one of claims 10 to 12, Both the information of the second result output device and the information of the first result output device are output and displayed on the first result output device. A monitoring and diagnostic method for a device, characterized by the above.

14. The monitoring and diagnostic method for a device according to any one of claims 10 to 12, The monitoring and diagnosis method of a device, characterized in that the first result output device and the second result output device are the same device, and the information presentation function as the first result output device and the information presentation function as the second result output device are switched and output and displayed by the user.

15. The monitoring and diagnosis method of a device according to any one of Claims 10 to 14, wherein the criterion for detecting the presence or absence of abnormality in the parameter correlation calculation unit is distinguished by the correlation change direction determination unit, and information is presented to the first result output device at a timing earlier than that of the second result output device. The monitoring and diagnosis method of a device is characterized by this.

16. A monitoring and diagnosis system for a device, wherein a measured value of the device measured by a sensor installed at a predetermined location of the device is transmitted to a monitoring and diagnosis device of the device via a network, and the device monitoring and diagnosis device detects a change in parameter correlation based on the transmitted measured value. The cause of the abnormality of the device is communicated to the owner of the device. The monitoring and diagnosis device of the device is the monitoring and diagnosis device of the device according to any one of Claims 1 to 8. The monitoring and diagnosis system of the device is characterized by this.

Images