EQUIPMENT CONDITION MONITORING DEVICE AND METHOD FOR EQUIPMENT CONDITION MONITORING

The equipment condition monitoring device addresses inaccuracies in existing methods by projecting current measurement data into a dimensionless space with corrected normal information, enabling precise equipment condition assessment.

DE112020007671B4Active Publication Date: 2024-11-28MITSUBISHI ELECTRIC CORP
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
DE112020007671
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-12-23
Publication Date
2024-11-28
Estimated Expiration
2040-12-23

AI Technical Summary

Technical Problem

Existing equipment condition monitoring methods struggle to accurately determine the condition of equipment due to variations in sensor data influenced by external environment changes and ideal vs. current operating environment discrepancies, leading to misidentification of equipment condition.

Method used

An equipment condition monitoring device that projects current measurement data into a dimensionless space where normal information is represented by a common form, correcting both normal and current measurement data based on their relationship in this space to estimate the equipment's condition accurately.

Benefits of technology

The device accurately detects equipment condition by minimizing the impact of outliers and environmental changes, ensuring precise recognition of the equipment's state through data correction and distribution estimation in a dimensionless space.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Equipment condition monitoring device (1), comprising: a data acquisition unit (11) to acquire portions of current measurement data (A), each indicating a status value of equipment; a normal information acquisition unit (12) to acquire portions of normal information, each indicating a normal state of the equipment; a position relationship acquisition unit (13) to capture the position relationship of each of the parts of the current measurement data (A) in a corresponding of the display forms (B) that display the respective parts of the normal information of the equipment; a projection unit (14) for projecting each of the parts of the current measurement data (A) into a dimensionless space based on the positional relationship between a corresponding display form (B) that displays the respective parts of the normal information and a corresponding part of the current measurement data (A), wherein the dimensionless space is a space in which the display forms (B) that display the respective parts of the normal information of the equipment are represented by a common form; a correction unit (15) for correcting both the normal information in dimensionless space and the current measurement data (A) in dimensionless space on the basis of a relationship between current measurement data (A), which indicates the normal state of the equipment in dimensionless space, and a display form (B), which indicates the normal information in dimensionless space; and a distribution estimation unit (16) to estimate a distribution of a state of the equipment based on the portions of the current measurement data (A) projected onto dimensionless space.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present disclosure relates to an equipment condition monitoring apparatus and an equipment condition monitoring method for monitoring a condition of equipment. Background to the state of the art

[0002] As a conventional equipment condition monitoring technique, for example, Patent Literature 1 describes a method for determining the maintenance timing of a vehicle air conditioner. In this method, the difference between the specific enthalpy at the start of compression and the specific enthalpy at the end of compression by the outdoor heat exchanger is compared with a first set value and a second set value greater than the first set value. If the specific enthalpy difference is greater than the second set value, it is determined that the air conditioner requires maintenance.

[0003] On the other hand, if the specific enthalpy difference is greater than the first set value and smaller than the second set value, a refrigeration cycle diagram of a Mollier diagram (hereinafter referred to as pH diagram (pressure-specific enthalpy)) is created based on sensor data acquired by various sensors in the outdoor heat exchanger. The refrigeration cycle diagram is displayed together with a refrigeration cycle diagram in the normal state of the air conditioner.

[0004] If the shape of the refrigerant cycle diagram in the pH diagram has changed from the corresponding normal state, it can be determined that the air conditioner's performance has deteriorated. If the shape deviation exceeds a certain value, it is determined that the air conditioner needs maintenance. It should be noted that the refrigerant cycle diagram in the normal state of the air conditioner is a display form of normal information that indicates the normal state of the equipment, and the sensor data used to create the refrigerant cycle diagram is the current measurement data of the equipment's condition value.

[0005] Further relevant disclosures can be found, for example, in patent literature 2 to 5. REFERENCE LISTPATENT LITERATURE Patent literature 1: JP 2015-92 121 A, Patent literature 2: DE 10 2004 023 824 A1, Patent literature 3: EP 0 775 335 B1, Patent literature 4: DE 699 10 800 T2 Patent literature 5: DE 11 2019 007 470 T5. SUMMARY OF THE INVENTIONTECHNICAL PROBLEM

[0006] The shape of the refrigerant cycle diagram of the air conditioner's pH diagram varies depending on the outside air temperature, even if the air conditioner's set temperature remains the same. Therefore, the refrigerant cycle diagram in the air conditioner's normal state has a different shape depending on the outside air temperature. Furthermore, the refrigerant cycle diagram to be compared with the refrigerant cycle diagram in the air conditioner's normal state is instantaneous data, which is created based on the sensor data acquired at the beginning of the maintenance time determination.

[0007] The method described in Patent Literature 1 monitors whether an equipment condition indicated by instantaneous data is normal or not to determine the need for equipment maintenance. Therefore, if an outlier occurs randomly in the instantaneous data, there is a possibility of misdetecting the equipment condition. The impact of random outliers in the sensor data can be mitigated based on a change trend in the equipment condition determined from a plurality of pieces of sensor data consecutively collected within a certain period of time (e.g., one month).

[0008] However, when the external environment of the equipment changes or the usage method of the equipment changes within a certain period of time, the normal state of the equipment changes accordingly. As a result, the normal information of the equipment is represented by a large number of display forms, which have different forms according to the changes in the normal state of the equipment. Therefore, it is difficult to understand the correspondence between the various pieces of sensor data and the normal information represented by a large number of display forms, and therefore the state of the equipment cannot be accurately recognized.

[0009] Furthermore, in the method described in Patent Literature 1, normal information (e.g., a refrigeration cycle diagram in a normal operating state of the air conditioner) is set for each operating state of the air conditioner based on a physical model that simulates the physical state of the air conditioner. Since the physical model simulates the physical state of the equipment operating in the predetermined ideal operating environment, the normal information of the equipment operating in the ideal operating environment is calculated based on the physical model.

[0010] Because the ideal operating environment of the equipment is generally different from the current operating environment of the equipment, an error will occur between the normal information of the equipment in the current operating environment and the normal information assuming the ideal operating environment of the equipment. Therefore, there is a deviation corresponding to the error between the sensor data acquired by the equipment in the current operating environment and the normal information assuming the ideal operating environment of the equipment. This raises the problem that the condition of the equipment cannot be accurately detected when comparing them.

[0011] Independent claims 1, 2 and 6 solve the problem described above and aim to obtain an equipment condition monitoring device and an equipment condition monitoring method capable of accurately detecting a condition of the equipment. SOLUTION TO THE PROBLEM

[0012] An equipment condition monitoring device according to the present disclosure includes: a data acquisition unit for acquiring the pieces of actual measurement data each indicating a condition value of the equipment; a normal information acquisition unit for acquiring pieces of normal information each indicating a normal condition of the equipment; a positional relationship acquisition unit for acquiring a positional relationship of each of the pieces of actual measurement data in a corresponding one of the display forms indicating the respective pieces of normal information of the equipment;a projection unit for projecting each of the pieces of current measurement data into a dimensionless space based on the positional relationship between a corresponding one of the display forms displaying the respective pieces of normal information and a corresponding one of the pieces of current measurement data, the dimensionless space being a space in which the display forms displaying the respective pieces of normal information of the equipment are represented by a common form; a correction unit for correcting both the normal information in the dimensionless space and the current measurement data in the dimensionless space based on a relationship between current measurement data displaying the normal state of the equipment in the dimensionless space and a display form displaying the normal information in the dimensionless space;and a distribution estimation unit for estimating a distribution of a state of the equipment based on the pieces of the current measurement data projected onto the dimensionless space; ADVANTAGEOUS EFFECTS OF THE INVENTION

[0013] According to the present disclosure, the pieces of current measurement data each indicating the state value of the equipment are projected into the dimensionless space, in which the display forms displaying the respective pieces of normal information each indicating the normal state of the equipment are represented by the common form, and the distribution of the equipment state is estimated based on the pieces of current measurement data projected into the dimensionless space. Furthermore, both the normal information in the dimensionless space and the current measurement data in the dimensionless space are corrected based on the relationship between the current measurement data indicating the normal state of the equipment in the dimensionless space and the display form displaying the normal information in the dimensionless space.As a result, the equipment condition monitoring device according to the present disclosure can accurately detect the condition of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a block diagram showing the configuration of an equipment condition monitoring device according to a first embodiment. Fig. 2 is a flowchart showing a method for equipment condition monitoring according to the first embodiment. Fig. 3A is a view illustrating a relationship between a plurality of display forms displaying a respective plurality of pieces of normal information of the equipment and a plurality of pieces of sensor data of the equipment, and Fig. 3B is a view illustrating a dimensionless space onto which a plurality of pieces of sensor data of the equipment are projected. Fig. Figure 4 is an explanatory diagram that provides an overview of the correction processing of the normal information of the equipment. Fig. Figure 5 is a flowchart illustrating the calculation of the positional relationship. Fig. 6 is a view illustrating a relationship between a display form showing normal information of the equipment and sensor data acquired by the equipment. Fig. 7 is a view illustrating the display form displaying the normal information of the equipment, the sensor data acquired by the equipment, and a center point of the display form displaying the normal information of the equipment. Fig. 8 is a view illustrating a positional relationship between the display form showing the normal information of the equipment and the sensor data acquired by the equipment. Fig. Figure 9 is a flowchart illustrating the projection processing of sensor data onto a dimensionless space. Fig. Figure 10 is a view showing an example of dimensionless space. Fig. 11 is a view that maps the sensor data to the dimensionless space of Fig. 10 projected. Fig. Figure 12 is a view showing an overview of the correction processing of sensor data of the equipment. Fig. Figure 13 is a flowchart illustrating the projection processing and correction processing of sensor data of the equipment. Fig. 14 is a view showing a positional relationship of sensor data of the equipment in dimensionless space. Fig. 15A is a view illustrating a distribution of sensor data projected onto the dimensionless space, and Fig. 15B is an explanatory diagram illustrating an overview of the estimation processing of a state of the equipment based on the distribution of the sensor data projected onto the dimensionless space. Fig. 16A is a block diagram illustrating a hardware configuration for implementing the functions of the plant condition monitoring device according to the first embodiment, and Fig. 16B is a block diagram illustrating a hardware configuration for executing software for implementing the functions of the equipment condition monitoring device according to the first embodiment. DESCRIPTION OF THE EMBODIMENTSFirst embodiment.

[0014] Fig. 1 is a block diagram showing a configuration of an equipment condition monitoring device 1 according to a first embodiment. In Fig. 1, the equipment condition monitoring device 1 is a device that monitors the condition of the equipment based on a comparison result of current measurement data indicating a condition value of the monitored equipment with a display form that displays normal information of the equipment in a dimensionless space. The normal information indicating a normal condition of the equipment is, for example, a range of a condition value indicating a normal operating condition of the equipment, and is determined for each operating condition of the equipment.

[0015] As in Fig. As shown in Figure 1, the equipment condition monitoring device 1 includes a data acquisition unit 11, a normal information acquisition unit 12, a positional relationship acquisition unit 13, a projection unit 14, a correction unit 15, a distribution estimation unit 16, and an output unit 17. The equipment whose operating state is monitored by the equipment condition monitoring device 1 is equipped with sensors for detecting various equipment condition values. The equipment to be monitored is, for example, an air conditioner.

[0016] The data acquisition unit 11 acquires current measurement data indicating a condition value of the equipment to be monitored. For example, the data acquisition unit 11 acquires current measurement data for a specific period to monitor the condition of the equipment. The period for monitoring the condition of the equipment is, for example, approximately one month. The current measurement data is, for example, sensor data acquired by the sensors installed in the equipment. If the equipment is an air conditioner, the sensors are, for example, pressure sensors and temperature sensors installed at various locations on the air conditioner. Based on the sensor data acquired by these sensors, the normal information of the air conditioner is determined for each operating state of the air conditioner.

[0017] The normal information acquisition unit 12 acquires the normal information of the equipment to be monitored. If the equipment to be monitored is an air conditioner, the normal information is, for example, a refrigerant cycle of a pH diagram in the normal state of the air conditioner. The normal information acquisition unit 12 calculates, for example, the refrigerant cycle in the normal state of the air conditioner using the sensor data in the normal operating state acquired by the pressure sensor and the temperature sensor in the air conditioner. That is, the "acquisition" of the normal information performed by the normal information acquisition unit 12 includes not only reading and acquiring the normal information stored in a storage device, but also calculating and acquiring the normal information using current measurement data indicating a state value of the equipment.

[0018] For example, a display format that shows the normal information of the air conditioner is a refrigerant cycle diagram in a normal operating state of the air conditioner. It should be noted that the refrigerant cycle diagram of the air conditioner changes depending on the outside air temperature, so the refrigerant cycle diagram that shows the normal information of the air conditioner has a large number of forms corresponding to different outside air temperatures. In addition, the sensors present in the air conditioner collect a large number of pieces of sensor data for each operating state of the air conditioner.For this reason, when a large number of pieces of sensor data are simply compared with normal information presented by a large number of display forms, it is difficult to detect a corresponding relationship between the distribution of sensor data and normal information, and therefore it is difficult to accurately detect the condition of the equipment.

[0019] The positional relationship acquisition unit 13 acquires a positional relationship of current measurement data in the display form of the normal information of the equipment. For example, if the display form that displays the normal information of the equipment is a polygon, the positional relationship acquisition unit 13 calculates, as the positional relationship of the current measurement data in the polygon, a distance between a center point of the polygon and the current measurement data, and an angle formed by a straight line passing through the center point of the polygon and the current measurement data and a side of the polygon that intersects the straight line. The center point of the polygon is, for example, the center of gravity.

[0020] Note that the “acquisition” of the positional relationship performed by the positional relationship acquisition unit 13 also includes reading and acquiring positional relationship data stored in the storage device in addition to calculating the positional relationship using the current measurement data and the normal information.

[0021] The projection unit 14 projects a plurality of pieces of current measurement data into a dimensionless space based on the positional relationships between a plurality of display forms displaying a respective plurality of pieces of normal information of the equipment and the plurality of pieces of current measurement data. The dimensionless space is a space in which a plurality of display forms displaying a respective plurality of pieces of normal information are represented by a common shape, and in which a point (dimensionless) representing a position of current measurement data in the display form displaying the normal information is arranged (projected).

[0022] The common shape is, for example, a circle with a radius of 1. The projection unit 14 treats the centers of the display forms of the plurality of pieces of normal information corresponding to the plurality of pieces of current measurement data as the center of the circle, and thereby converts the positional relationship between the center of the display form of the normal information and the current measurement data into the positional relationship with the center of the circle. The converted positional relationship includes, for example, distance information and angle information. The projection unit 14 arranges points of the plurality of pieces of current measurement data in the dimensionless space including the circle. The representation of the current measurement data in the dimensionless space is a point of a dimensionless value indicating the positional relationship of the current measurement data with respect to the display form indicating the normal information of the equipment.Furthermore, the distribution of points in dimensionless space corresponds to the distribution of areas corresponding to the condition values ​​of the equipment.

[0023] The correction unit 15 corrects both the normal information in dimensionless space and the current measurement data in dimensionless space based on the relationship between the current measurement data indicating the normal state of the equipment in dimensionless space and the display form displaying the normal information in dimensionless space, or one of them. For example, if the display form displaying the normal information of the equipment is a circle with a radius of 1 in dimensionless space, the circle serves as a threshold for determining whether the operating state of the equipment is a normal state or an abnormal state. In addition, the current measurement data acquired by the equipment in the normal operating state are most distributed at the center of the circle with a radius of 1, which indicates the range of the state value of the equipment in the normal state.The center of the circle corresponds to the origin of the coordinate system in dimensionless space (hereinafter referred to as the origin of dimensionless space). In other words, the deviation between the center of the distribution of the current measurement data projected onto dimensionless space and the center of the circle with radius 1 corresponds to an error between the normal information in the current operating environment and the normal information assuming the ideal operating environment.

[0024] Therefore, the correction unit 15 calculates the shortest distance between the center position of the distribution of the current measurement data projected onto the dimensionless space and the circle with a radius of 1, and corrects the normal information and the current measurement data so that the display form displaying the normal information becomes a circle with the calculated shortest distance as the radius. This corrects the relationship between the equipment's normal information and the current measurement data, and the equipment's condition can be accurately recognized by comparing these data.

[0025] The distribution estimation unit 16 estimates the distribution of the equipment's states based on a plurality of pieces of current measurement data projected onto dimensionless space. For example, the distribution estimation unit 16 estimates the distribution of points in dimensionless space, that is, the distribution of the equipment's state values, by performing Gaussian mixture model estimation on the points of the plurality of pieces of current measurement data arranged in dimensionless space. The method for estimating the distribution of the equipment's state values ​​only needs to be able to estimate the probability density distribution of the points in dimensionless space, and it can use, for example, a maximum likelihood method, Bayesian estimation, or an EM algorithm.

[0026] The output unit 17 outputs data that can be used to monitor the condition of the equipment. The data to be used to monitor the condition of the equipment is, for example, a dimensionless space in which a plurality of display forms each displaying a plurality of pieces of normal information are represented by a common form, and in which the distribution of the condition values ​​of the equipment is arranged with respect to a range indicated by the common form. The output unit 17 causes a display device, for example, to display the dimensionless space. The output unit 17 may be a component included in an external device provided separately from the equipment condition monitoring device 1.In this case, the equipment condition monitoring device 1 does not include an output unit 17, and the distribution of the condition values ​​of the equipment is displayed on a display device included in the external device by the output unit 17 included in the external device.

[0027] Furthermore, the equipment condition monitoring device 1 may include only the correction unit 15 and the output unit 17, and the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, and the distribution estimation unit 16 may be components of an external device provided separately from the equipment condition monitoring device 1.

[0028] In this case, the correction unit 15 in the equipment condition monitoring device 1 corrects both the normal information in the dimensionless space and the current measurement data in the dimensionless space based on the relationship between the current measurement data indicating the normal state of the equipment in the dimensionless space and the display form displaying the normal information in the dimensionless space, which relationship is acquired by the projection unit 14 in the external device. Then, the output unit 17 causes a display device included in the equipment condition monitoring device 1 or a display device included in the external device to display the dimensionless space in which the distribution of the states of the equipment is projected onto a common form displaying each of a plurality of pieces of the normal information of the equipment.

[0029] Fig. 2 is a flowchart illustrating an equipment condition monitoring method according to the first embodiment, and illustrating the operation of the equipment condition monitoring device 1. The current measurement data indicating the condition value of the equipment is the following sensor data. The data acquisition unit 11 acquires sensor data indicating a condition value of the equipment to be monitored (step ST1). For example, by continuously or periodically receiving sensor data sequentially acquired by the sensor provided in the equipment, the data acquisition unit 11 acquires a time series of sensor data during a certain period (hereinafter referred to as a monitoring period) for monitoring the equipment, that is, time series data on a condition value of the equipment.Note that the sensor data may be stored in a storage device, and the data acquisition unit 11 may acquire the sensor data from the storage device.

[0030] Subsequently, the normal information acquisition unit 12 acquires normal information indicating the normal state of the equipment to be monitored (step ST2). For example, if the equipment to be monitored is an air conditioner and the display form displaying the normal information is the refrigerant cycle diagram in the pH diagram in the normal operating state of the air conditioner, the normal information acquisition unit 12 calculates the data of the refrigerant cycle diagram in the normal operating state of the air conditioner using the sensor data acquired from the air conditioner in the normal operating state by the pressure sensor and the temperature sensor in the air conditioner.Alternatively, the refrigeration cycle map data in the normal operating state of the air conditioner may be stored in a storage device, and the normal information acquiring unit 12 may sequentially acquire the refrigeration cycle map data from the storage device according to the operating state in the monitoring period.

[0031] The positional relationship acquisition unit 13 acquires a positional relationship of the sensor data in the display form indicating the normal information of the equipment to be monitored (step ST3). Fig. 3A is a view illustrating a relationship between a plurality of display forms, each displaying a plurality of pieces of normal information of the equipment, and a plurality of sensor data of the equipment, and illustrating a relationship between sensor data (1) and sensor data (2) acquired by the sensors provided in the equipment. For example, the positional relationship acquisition unit 13 creates the Fig. 3A using the current measurement data acquired by the data acquisition unit 11 and the normal information acquired by the normal information acquisition unit 12.

[0032] In the Fig. In the view shown in Figure 3A, diagrams of a plurality of pieces of sensor data A are arranged, each including the sensor data (1) and the sensor data (2), and refrigeration cycle diagrams B are displayed, indicating a plurality of pieces of normal information corresponding to these respective points. As shown in Fig. As shown in Figure 3A, the shape of the refrigerant cycle diagram B varies depending on the external environment or the operating state of the air conditioner, even if the operating states specified from the sensor data (1) and the sensor data (2) are the same. When the refrigerant cycle diagram B of the normal information varies, the correspondence relationship between the point of the sensor data A and the refrigerant cycle diagram B of the normal information becomes complicated, as shown in Fig. 3A, and the condition of the equipment cannot therefore be accurately determined.

[0033] Therefore, the equipment condition monitoring device 1 projects a plurality of pieces of current measurement data acquired by the equipment into a dimensionless space in which the plurality of display forms displaying the respective plurality of pieces of normal information are represented by a common form. This allows the equipment condition monitoring device 1 to recognize multiple states of the equipment from the common display form. The positional relationship acquisition unit 13 acquires, as information for projecting the current measurement data onto the dimensionless space, a positional relationship of the current measurement data in the display form indicating the normal information.This positional relationship includes, for example, information indicating a distance between the center of the display form indicating the normal information and the current measurement data, and information indicating an angle formed by a straight line passing through the center of the display form indicating the normal information and the current measurement data and a side of the display form indicating the normal information that intersects the straight line.

[0034] The projection unit 14 projects a plurality of sensor data acquired from the equipment into the dimensionless space based on positional relationships between the plurality of display forms displaying the respective plurality of pieces of the normal information of the equipment and the plurality of sensor data (step ST4). Fig. Figure 3B is a view illustrating a dimensionless space onto which a variety of pieces of the equipment's sensor data are projected. In Fig. 3B, for example, the projection unit 14 treats the center point of each of the display forms displaying the plurality of pieces of normal information corresponding to the plurality of sensor data as the center point 21 of a circle 20 in dimensionless space, and thereby converts the positional relationships between the centers of the display forms displaying the pieces of normal information and the pieces of sensor data into positional relationships with the center point 21 of the circle 20, thereby arranging the points of the plurality of pieces of sensor data in the circle 20. This results in a distribution 22 of the sensor data in the circle 20.

[0035] The position of the sensor data point in dimensionless space is a position corresponding to the relationship between the sensor data before projection and the display form that displays the normal information. It should be noted that a Fig. 3B, component (1) is a component corresponding to a change in the sensor data (1), and component (2) is a component corresponding to a change in the sensor data (2).

[0036] The correction unit 15 corrects the normal information and the sensor data based on a relationship between the sensor data and the normal information (step ST5). Fig. Figure 4 is an explanatory diagram that provides an overview of the correction processing of the equipment's normal information. Fig. 4 is the display form that shows the normal information of the equipment, a circle 20 with a radius of 1 in dimensionless space. The center point 21 of circle 20 is the origin of the coordinate system of dimensionless space. The sensor data acquired by the equipment in normal operating conditions are most densely distributed in the center 21 of circle 20.

[0037] As shown in the view on the left in Fig. 4 However, in dimensionless space, before correcting the normal information of the equipment, a center point 23 of a distribution 22 of the sensor data deviates from the center point 21 of the circle 20. A deviation amount between the center point 21 of the circle 20 and the center point 23 of the distribution 22 of the sensor data in the coordinate system of dimensionless space corresponds to an error between the normal information in the current operating environment and the normal information assuming the ideal operating environment.

[0038] The correction unit 15 corrects the normal information so that there is no deviation between the center point 21 of the circle 20 and the center point 23 of the distribution 22 of the sensor data. That is, the normal information of the equipment is corrected to be represented by a circle 20A centered on the center point 23 of the distribution 22 of the sensor data. The correction unit 15 searches for, for example, a nearest point 24 between the center point 23 of the distribution 22 of the sensor data and the circle 20. The position of the center point 23 of the distribution 22 of the sensor data is, for example, a position corresponding to a mean or a median of coordinate values ​​in the X-axis and Y-axis directions of points of the sensor data arranged in dimensionless space.

[0039] The correction unit 15 calculates a distance r' between the found nearest point 24 and the center 23 of the distribution 22 of the sensor data and corrects the normal information to be represented by the circle 20A with a radius of r', as shown in the view on the right in Fig. 4. The correction unit 15 can then correct the deviation in the distribution 22 of the sensor data to 1 / r' times. Thus, the correction unit 15 creates a new distribution 22A in the circle 20A, in which the distance between the center 23 of the distribution 22 and each point is multiplied by 1 / r'. This corrects the current measurement data in addition to the normal information from the equipment.

[0040] The distribution estimation unit 16 estimates the distribution of the states of the equipment based on the plurality of pieces of the sensor data projected onto the dimensionless space (step ST6). For example, the distribution estimation unit 16 estimates the distribution 22A in the Fig. 4 by performing Gaussian mixture model estimation at a plurality of points arranged in the dimensionless space. In the distribution 22A, the color or hue is determined depending on the number of points corresponding to the sensor data, with the number of points corresponding to the sensor data being largest in an area where the color is darkest and decreasing with increasing brightness. The output unit 17 displays the dimensionless space in which the distribution 22A is located on the display device.

[0041] Fig. 5 is a flowchart illustrating the calculation of the positional relationship and details of the processing in step ST3 of Fig. 2. The positional relationship acquisition unit 13 displays a two-dimensional coordinate plane in which the points of the equipment sensor data acquired by the data acquisition unit 11 and the display forms of the equipment normal information acquired by the normal information acquisition unit 12 are arranged (step ST1a).

[0042] Fig. Figure 6 is a view illustrating a relationship between a display form showing normal information of the equipment and sensor data acquired by the equipment. Fig. 6, the sensor data 31 is data indicating a state value of the equipment determined from the sensor data (1) and the sensor data (2) acquired by the equipment to be monitored. A display shape 32 is normal information for the operating state of the equipment displayed by the sensor data 31 and is a quadrangular shape with vertices 33 to 36.

[0043] The positional relationship acquisition unit 13 calculates a center point 37 of the display form 32 (step ST2a). Fig. Fig. 7 is a view of the display form 32, which displays the normal information of the equipment, the sensor data 31 acquired by the equipment, and the center 37. For example, the positional relationship acquisition unit 13 calculates the center of gravity of the outer shape of the display form 32 as the center 37. Since the display form 32 in Fig. 7 is a quadrilateral, the positional relationship acquisition unit 13 calculates an intersection point of a diagonal line connecting the vertex 33 and the vertex 35 and a diagonal line connecting the vertex 34 and the vertex 36 as the center point 37.

[0044] Fig. Fig. 8 is a view showing a positional relationship between the display form 32, which displays the normal information of the equipment, and the sensor data 31 acquired by the equipment. As shown in Fig. 8, the positional relationship acquisition unit 13 calculates in addition to a distance r a between the sensor data 31 and the center point 37 of the display form 32 an angle θ a, which is formed by a straight line A' passing through the sensor data 31 and the center point 37 of the display form 32, and a side C of the display form 32 intersecting the straight line A' (step ST3a). Subsequently, the positional relationship acquisition unit 13 calculates a distance r n from the center point 37 of the display form 32 to an intersection point Ca between the straight line A and the side C of the display form 32.

[0045] The positional relationship acquisition unit 13 calculates a relative ratio r R (= r a / r n ) of the distance r a to the distance r a , which is the length of the line A. Then, the positional relationship acquisition unit 13 calculates the angle θ a , which is formed by the line A and the side C. The relative ratio r R and the angle θ calculated in this way aare output to the projection unit 14 as data indicating the positional relationship of the sensor data 31 with respect to the center point 37 of the display form 32.

[0046] Fig. 9 is a flowchart illustrating the projection processing of sensor data onto a dimensionless space and details of the processing in step ST4 of Fig. 2 shows. Fig. Figure 10 is a view showing an example of dimensionless space. The dimensionless space can be a space in which a circle 20 with a radius of 1, as in Fig. 10. Assuming that the angle to the axis of component (1) is θ, a point on circle 20 is represented by (1, θ). Component (1) is a component corresponding to a change in sensor data (1) of Fig. 6, and the component (2) is a component that corresponds to a change in the sensor data (2) of Fig. 6 corresponds.

[0047] The projection unit 14 projects the sensor data (current measurement data) 31 into the dimensionless space based on the positional relationship of the sensor data 31 in the display form 32 (step ST1b). Fig. 11 is a view projecting the sensor data 25 onto the dimensionless space of Fig. 10. For example, the projection unit 14 treats the center point 37 of the Fig. 8 shown display form 32 as the center 21 of the circle 20 in the Fig. 10 dimensionless space. By defining it by a circle 20 with radius r n replaced and the relative ratio r R between the distance between the center point 37 of the display form 32 and the sensor data 31 and the angle θ a, which is formed by the straight line A passing through the sensor data 31 and the side C of the display form 32, the projection unit 14 represents a point of the sensor data 25 in dimensionless space. As in Fig. 11, the representation of the sensor data 25 is divided into a point (r R , θ a ) and arranged in dimensionless space.

[0048] The projection unit 14 checks whether all pieces of the sensor data 31, from which the positional relationships to the respective display forms 32 have been acquired, have been projected into the dimensionless space or not (step ST2b). If there is sensor data 31 that has not yet been projected into the dimensionless space (step ST2b; NO), the processing returns to step ST1b, and the center points 37 of the plurality of display forms 32 are sequentially treated as the center point 21 of the circle 20, and thereby the sensor data 31 is projected into the dimensionless space. If there is no sensor data 31 that has not been projected into the dimensionless space (step ST2b; YES), the processing ends. Fig. 9.

[0049] The correction unit 15 corrects both the normal information in the dimensionless space and the sensor data in the dimensionless space based on the relationship between the sensor data indicating the normal state of the equipment in the dimensionless space and the display form indicating the normal information in the dimensionless space, or one of them. Fig. Figure 12 is a view that provides an overview of the correction processing of the equipment's sensor data. Assuming that the equipment is not damaged, the distribution of the sensor data A obtained by the equipment in normal operation is distributed near the center point 37 of the normal information. However, if the individual difference of the equipment or the information required for calculating the normal information is not sufficiently obtained, an error will occur in the equipment's normal information. This will result in, as shown in Fig. 12, a deviation 40 between the sensor data A and the center point 37 of the display form B.

[0050] As in Fig. As shown in Figure 12, when similar display forms B are extracted from a plurality of display forms B (e.g., a refrigeration cycle diagram) each displaying a plurality of pieces of normal information, the operating states of the equipment corresponding to the similar display forms B are similar. A plurality of pieces of sensor data A indicating similar operating states tend to be distributed at close positions in the display form B displaying the normal information of the equipment. The correction unit 15 utilizes this tendency to estimate the deviation 40 between the sensor data A and the center point 37 of the display form B, thereby correcting the sensor data A and the display form B.

[0051] Fig. Fig. 13 is a flowchart illustrating the projection processing and the correction processing of the sensor data of the equipment and showing details of the processing of steps ST4 and ST5 in Fig. 2 illustrates. Fig. Figure 14 is a view showing a positional relationship of the sensor data (hereinafter referred to as actual measurement data) of the equipment in the dimensionless space. Fig. For example, in the dimensionless space shown in Figure 14, the projection unit 14 has a circle 20 with a radius of 1, which displays the normal information of the equipment, and a point of the current measurement data (step ST1c).

[0052] The correction unit 15 calculates a position (x, y) of the point of the current measurement data of the equipment in the dimensionless space whose origin is the center 21 of the circle 20 (step ST2c). Note that the position (x, y) of the point of the current measurement data of the equipment is a position indicating a deviation from the origin of the dimensionless space. The position in the dimensionless space can be represented in a polar coordinate format.

[0053] Next, the correction unit 15 determines whether the current measurement data for which the position (x, y) was calculated is data obtained during an initial learning period (step ST3c). The initial learning period is a period for learning a relationship between the position coordinates of the display forms displaying normal information in dimensionless space and the position coordinates of a point of a plurality of pieces of current measurement data in dimensionless space obtained by the equipment in a normal state. The initial learning period is set in advance, for example, by the user.

[0054] If the current measurement data is data acquired during the initial learning period (step ST3c; YES), the correction unit 15 learns the deviation of the current measurement data from the origin based on the relationship between the normal information and the current measurement data (step ST4c). For example, the correction unit 15 acquires a learner that has learned the correlation between the position (x, y) of the point of the current measurement data acquired by the equipment in the normal state within the initial learning period and the position coordinates of the display form that displays the equipment's normal information in dimensionless space. In this case, the learning data includes the position of the points of the current measurement data acquired by the equipment in the normal state and the position of the display form that displays the equipment's normal information in dimensionless space.

[0055] The learner uses as input position coordinates of a display form that displays normal information of the equipment in the dimensionless space, and estimates a position (x', y') of estimated data in the dimensionless space, where the estimated data indicates a normal state value in different operating states of the equipment. The position (x', y') of the estimated data is a position indicating a deviation from the origin of the dimensionless space. In addition, the position coordinates of the display forms that display the normal information of the equipment in the dimensionless space, for example, the position coordinates of the vertices 33 to 36 of the quadrilateral form that displays the normal information of the Fig. 6. The learner can perform the estimation using multiple regression analysis or a neural network.

[0056] In a case where the learner is acquired during the initial learning period, or in a case where the current measurement data is not data acquired during the initial learning period and the learner is acquired because the initial learning period has elapsed (step ST3c; NO), the correction unit 15 estimates the position (x', y') of the estimation data using the learner (step ST5c). For example, the correction unit 15 outputs the position coordinates of the vertices 33 to 36 of the quadrilateral shape containing the normal information of the Fig. 6, to the learner, thereby estimating the position (x', y') of the estimation data in dimensionless space.

[0057] The correction unit 15 corrects the position (x, y) of the point of the current measurement data obtained after the initial learning period by using the position (x', y') of the estimated data corresponding to the state indicated by the current measurement data (step ST6c). For example, the correction unit 15 subtracts the position (x', y') of the estimated data from the position (x, y) of the point of the current measurement data of the equipment in dimensionless space, thereby correcting the current measurement data so that the deviation from the origin of the dimensionless space is reduced. As a result, the position (x-x', y-y') of the current measurement data after the correction is obtained. In this way, the current measurement data acquired by the equipment after the initial learning period is corrected.

[0058] It should be noted that the learning data for the learner can be actual measurement data acquired by the equipment in the normal state. That is, by using the actual measurement data acquired by the equipment in the normal state as learning data, the learner learns the correlation between the actual measurement data acquired by the equipment in the normal state and the position of the display form that displays the normal information of the equipment in the dimensionless space. Also in this way, the learner can be obtained which takes as input position coordinates of a display form that displays normal information of the equipment in the dimensionless space and which estimates the position (x', y') of the estimated data in the dimensionless space, where the estimated data indicates the normal state value in different operating states of the equipment.

[0059] The distribution estimation unit 16 estimates the distribution 22 of the current measurement data in the Fig. 3B by, for example, performing the Gaussian mixture model estimation at the points of the plurality of pieces of the current measurement data whose positions in the dimensionless space have been corrected by the correction unit 15. Fig. Figure 15A is a view showing the distribution 22 of the sensor data projected onto dimensionless space. In the distribution 22 of the sensor data, the number of sensor data points is largest in an area 28 with the darkest color and decreases with increasing color brightness.

[0060] In Fig. 15A, component (1) is a component corresponding to a change in sensor data (1), and component (2) is a component corresponding to a change in sensor data (2). In the distribution 22, the region 28 has the most points of the sensor data, and the sensor data in the region 28 has a dominant influence on the feature of the distribution 22. The circle 20 is a display shape indicating a normal state of the equipment indicated by sensor data including the sensor data (1) and the sensor data (2).

[0061] The distribution estimation unit 16 determines whether the condition of the monitored equipment is close to the normal state based on the positional relationship between the area 28 of the distribution 22 and the circle 20. Since the area 28 of the distribution 22 is located at a position away from the center point 21 of the circle 20, it can be seen that the condition of the equipment is outside the normal state, that is, the condition of the equipment is deteriorating. As described above, the equipment condition monitoring device 1 can indicate the change trend of the equipment condition during the monitoring period, and even if an outlier occurs randomly in the sensor data, its influence is reduced.

[0062] Fig. Figure 15B is an explanatory diagram that provides an overview of the processing of estimating the state of the equipment based on the distribution of the sensor data projected onto the dimensionless space. In Fig. 15B, a distribution 22A is a distribution of the equipment states estimated by the distribution estimation unit 16, and estimated based on sensor data acquired in a monitoring period (1). A distribution 22B is a distribution in the same state as the distribution 22A, but a distribution estimated based on sensor data acquired in a monitoring period (2) subsequent to the monitoring period (1). Furthermore, the outer shapes of the distribution 22A and the distribution 22B are elliptical.

[0063] The state of the equipment to be monitored changes over time, so that distribution 22A transitions into distribution 22B. At this time, the transition from the center point 23A of distribution 22A to the center point 23B of distribution 22B can be determined by (r c , θ c ). The distance r cis a distance between the center 23A and the center 23B, and the angle θ c is an angle formed by a line segment connecting the center 23A and the center 23B at the center 23A. Similarly, the transition from the point 27A on the major axis of the distribution 22A to the point 27B on the major axis of the distribution 22B can be described by (r m , θ m ), and the transition from point 28A on the minor axis of distribution 22A to point 28B on the minor axis of distribution 22B can be represented by (r s , θ s ) can be displayed.

[0064] The distribution estimation unit 16 specifies the positional relationship between the distribution of the sensor data projected onto the dimensionless space and, for example, the Fig. 15A based on the (r, θ) representing the transition of the distribution of the current measurement data projected onto the dimensionless space, and determines whether the condition of the equipment to be monitored has deteriorated or not based on the specified positional relationship. In this way, by quantifying the temporal transition of the distribution estimated by the distribution estimation unit 16, the deterioration of the condition of the equipment to be monitored can be accurately determined.

[0065] In the equipment condition monitoring device 1, the display shape that displays the normal information of the equipment in the dimensionless space may be a circular shape or a polygonal shape.

[0066] The functions of the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17 in the equipment condition monitoring device 1 are realized by a processing circuit. That is, the equipment condition monitoring device 1 includes a processing circuit for executing the processing from step ST1 to step ST6 in Fig. 2. The processing circuitry may be dedicated hardware or a central processing unit (CPU) that executes a program stored in memory.

[0067] Fig. 16A is a block diagram showing a hardware configuration that implements the functions of the equipment condition monitoring device 1. Fig. 16B is a block diagram showing a hardware configuration for executing software that implements the functions of the equipment condition monitoring device 1. In Fig. 16A and Fig. 16B, an input interface 100 is an interface that transmits sensor data output from the sensor provided in the equipment to be monitored to the equipment condition monitoring device 1. Furthermore, an output interface 101 is an interface that transmits information output from the equipment condition monitoring device 1.

[0068] In a case where the processing circuit is a processing circuit 102 of dedicated hardware, as in Fig. 16A, the processing circuit 102 corresponds, for example, to a single circuit, a composite circuit, a programmed processor, a parallel-programmed processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination thereof. The functions of the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17 in the equipment condition monitoring device 1 can be implemented by separate processing circuits, or these functions can be implemented collectively by one processing circuit.

[0069] In a case where the processing circuitry is Fig. 16B, the functions of the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17 in the equipment condition monitoring device 1 are implemented by software, firmware, or a combination of software and firmware. Note that software or firmware is written as a program and stored in a working memory 104.

[0070] The processor 103 reads and executes the program stored in the memory 104 to implement the functions of the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17 in the equipment condition monitoring device 1. The equipment condition monitoring device 1 includes, for example, a memory 104 for storing a program which, when executed by the processor 103, is used to execute the processing steps ST1 to ST5 in the Fig.2. These programs cause a computer to execute operations or methods performed by the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17. The memory 104 may be a computer-readable storage medium that stores a program that causes a computer to operate as the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17.

[0071] Examples of the memory 104 correspond to a non-volatile or volatile semiconductor memory such as a random access memory (RAM), a read-only memory (ROM), a flash memory, an erasable programmable read-only memory (EPROM) or an electrical EPROM (EEPROM), a magnetic disk, a flexible disk, an optical disk, a compact disc, a minidisk and a DVD.

[0072] Some of the functions of the data acquisition unit 11, the normal information acquisition unit 12, the positional relationship acquisition unit 13, the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17 in the equipment condition monitoring device 1 can be implemented by dedicated hardware, and other parts can be implemented by software or firmware. For example, the functions of the data acquisition unit 11, the normal information acquisition unit 12, and the positional relationship acquisition unit 13 are performed by the processing circuit 102, which is dedicated hardware, and the functions of the projection unit 14, the correction unit 15, the distribution estimation unit 16, and the output unit 17 are performed by the processor 103, which reads and executes the program stored in the memory 104.As described above, the processing circuitry may implement the above functions through hardware, software, firmware, or a combination thereof.

[0073] As described above, the equipment condition monitoring device 1 according to the first embodiment projects a plurality of pieces of current measurement data indicating a condition value of the equipment into a dimensionless space in which a plurality of display forms each displaying a plurality of pieces of normal information indicating a normal condition of the equipment are represented by a common form, and estimates the distribution of the condition of the equipment based on the plurality of pieces of current measurement data projected into the dimensionless space. Furthermore, both the normal information in the dimensionless space and the current measurement data in the dimensionless space are corrected based on the relationship between the current measurement data indicating the normal condition of the equipment in the dimensionless space and the display form displaying the normal information in the dimensionless space.This allows the equipment condition monitoring device 1 to accurately detect the condition of the equipment.

[0074] In the equipment condition monitoring device 1 according to the first embodiment, the positional relationship acquisition unit 13 calculates, as information indicating the positional relationship of the current measurement data in the display form displaying the normal information, a distance between a center point of the display form displaying the normal information and the current measurement data, and an angle formed by a straight line passing through the center point and the current measurement data and a side of the display form displaying the normal information, the straight line intersecting the side. This makes it possible to accurately acquire the information indicating the positional relationship of the current measurement data in the display form displaying the normal information.

[0075] In the equipment condition monitoring device 1 according to the first embodiment, the correction unit 15 includes the learner that estimates a position of data indicating a normal state value of the equipment in the dimensionless space using a position of a display form displaying the normal information in the dimensionless space as input. The correction unit 15 corrects the current measurement data using the position of the data estimated by the learner in such a manner as to reduce a deviation from an origin of the dimensionless space. Thus, it is possible to accurately correct the deviation of the current measurement data projected onto the dimensionless space from the origin of the dimensionless space.

[0076] It should be noted that any component of the embodiment may be changed or any component of the embodiment may be omitted. INDUSTRIAL APPLICABILITY

[0077] The equipment condition monitoring device according to the present disclosure can be used, for example, to monitor an operating condition of an air conditioner. LIST OF REFERENCE SYMBOLS

[0078] 1: Equipment condition monitoring device, 11: Data acquisition unit, 12: Normal information acquisition unit, 13: Positional relationship acquisition unit, 14: Projection unit, 15: Correction unit, 16: Distribution estimation unit, 17: Output unit, 20, 20A: Circle, 21: Center point, 22, 22A, 22B: Distribution, 23, 37: Center point, 23A, 23B: Center point, 24: Nearest point, 25, 31: Sensor data, 27A, 27B: Point on the major axis, 28: Region, 28A, 28B: Point on the minor axis, 32: Display shape, 33 to 36: Vertex, 40: Deviation, 100: Input interface, 101: Output interface, 102: Processing circuit, 103: Processor, 104: Memory, A: Sensor data, B: Display form, C: Display form page, C a : intersection point, θ a , θ c , θ m , θ s : angle, r a , r c , r m ,r n , r s : distance, r R : relative ratio r a to n, A': straight line, x, y: position, ST1-ST6, ST1a-ST3a, ST1b, ST2b, ST1c-ST6c: process steps

Claims

[1] Equipment condition monitoring device (1) comprising: a data acquisition unit (11) for acquiring pieces of current measurement data (A) each indicating a condition value of equipment; a normal information acquiring unit (12) for acquiring pieces of normal information each indicating a normal state of the equipment; a positional relationship acquiring unit (13) for acquiring a positional relationship of each of the pieces of the current measurement data (A) in a corresponding one of the display forms (B) displaying the respective pieces of the normal information of the equipment; a projection unit (14) for projecting each of the pieces of the current measurement data (A) into a dimensionless space based on the positional relationship between a corresponding one of the display forms (B) displaying the respective pieces of the normal information and a corresponding one of the pieces of the current measurement data (A), wherein the dimensionless space is a space in which the display forms (B) displaying the respective pieces of the normal information of the equipment are represented by a common shape; a correction unit (15) for correcting both the normal information in the dimensionless space and the current measurement data (A) in the dimensionless space on the basis of a relationship between current measurement data (A) indicating the normal state of the equipment in the dimensionless space and a display form (B) indicating the normal information in the dimensionless space; and a distribution estimation unit (16) for estimating a distribution of a state of the equipment based on the parts of the current measurement data (A) projected onto the dimensionless space. [2] An equipment condition monitoring device (1) configured to monitor a condition of equipment, the equipment condition monitoring device (1) comprising: an output unit (17) configured to output data for monitoring the condition of the equipment; and a correction unit (15) configured to correct current measurement data (A) indicating a state value of the equipment or normal information indicating a normal state of the equipment; wherein the output unit (17) is designed to output a dimensionless space in which display forms (B) displaying respective pieces of the normal information of the equipment are represented by a common form, and a distribution of the state of the equipment is projected onto the common form, and the correction unit (15) is configured to correct both the normal information in the dimensionless space and the current measurement data (A) in the dimensionless space based on a relationship between current measurement data (A) indicating the normal state of the equipment in the dimensionless space and a display form (B) indicating the normal information in the dimensionless space. [3] Equipment condition monitoring device (1) according to claim 2, further comprising: a data acquisition unit (11) configured to acquire parts of the current measurement data (A); a normal information acquiring unit (12) configured to acquire the pieces of normal information; a positional relationship acquisition unit (13) configured to acquire a positional relationship of each of the pieces of the current measurement data (A) in a corresponding one of the display forms (B) displaying the respective pieces of the normal information; a projection unit (14) configured to project each of the pieces of the current measurement data (A) into the dimensionless space based on the positional relationship between a corresponding one of the display forms (B) displaying the respective pieces of the normal information of the equipment and a corresponding one of the pieces of the current measurement data (A); and a distribution estimation unit (16) configured to estimate the distribution of the equipment state based on the parts of the current measurement data (A) projected onto the dimensionless space. [4] The equipment condition monitoring device (1) according to claim 1 or 3, wherein the positional relationship acquisition unit (13) is configured to acquire, as information indicating the positional relationship of each of the pieces of the current measurement data (A) in the corresponding one of the display forms (B) displaying the respective pieces of the normal information, a distance (r a ) between a center point (37) of the corresponding one of the display forms (B) displaying the respective parts of the normal information and the corresponding part of the current measurement data (A), and an angle (θ a) formed by a straight line (A') passing through the center point (37) and the corresponding part of the current measurement data (A) and a side (C) of the corresponding display form (B) displaying the respective parts of the normal information, the straight line (A') intersecting the side (C). [5] The equipment condition monitoring device (1) according to any one of claims 1 to 3, wherein the correction unit (15) comprises a learner for estimating a position of data indicating a normal state value of the equipment in the dimensionless space using a position of a display form (B) displaying the normal information in the dimensionless space as input, and correcting the current measurement data (A) using the position (x,y) of the data estimated by the learner in such a manner as to reduce a deviation from an origin (21) of the dimensionless space. [6] A method for monitoring the condition of equipment, the method comprising the steps of: Acquiring (ST1), by a data acquisition unit (11), current measurement data (A) each indicating a state value of equipment; Acquiring (ST2), by a normal information acquiring unit (12), normal information each indicating a normal state of the equipment; Acquiring (ST3), by a positional relationship acquiring unit (13), a positional relationship of each of the pieces of the current measurement data (A) in a corresponding display form (B) displaying the respective pieces of the normal information of the equipment; Projecting (ST4), by a projection unit (14), each of the pieces of the current measurement data (A) into a dimensionless space based on the positional relationship between a corresponding one of the display forms (B) displaying the respective pieces of the normal information and a corresponding one of the pieces of the current measurement data (A), wherein the dimensionless space is a space in which the display forms (B) displaying the respective pieces of the normal information of the equipment are represented by a common shape; Correcting (ST5), by a correction unit (15), both the normal information in the dimensionless space and the current measurement data (A) in the dimensionless space on the basis of a relationship between current measurement data (A) indicating the normal state of the equipment in the dimensionless space and a display form (B) indicating the normal information in the dimensionless space; and Estimating (ST6), by a distribution estimation unit (16), a distribution (22A) of an equipment state based on the parts of the current measurement data (A) projected onto the dimensionless space.

Citation Information

Patent Citations

  • device and method for assessing a quality class of an object to be inspected

    DE102004023824A1

  • DEVICE FOR MONITORING THE CONDITION OF A DEVICE AND METHOD FOR MONITORING THE CONDITION OF A DEVICE

    DE112019007470T5

  • Method and device for monitoring the operating status of an individual machine

    DE69910800T2

  • Method for monitoring multivariate processes

    EP0775335B1

  • Maintenance timing determination method for vehicular air conditioner and air conditioner

    JP2015092121A