CN109798970B - Abnormality detection device, abnormality detection method, abnormality detection system, and storage medium - Google Patents
Abnormality detection device, abnormality detection method, abnormality detection system, and storage medium Download PDFInfo
- Publication number
- CN109798970B CN109798970B CN201811248265.8A CN201811248265A CN109798970B CN 109798970 B CN109798970 B CN 109798970B CN 201811248265 A CN201811248265 A CN 201811248265A CN 109798970 B CN109798970 B CN 109798970B
- Authority
- CN
- China
- Prior art keywords
- unit
- abnormality detection
- machine
- abnormality
- vibration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
Provided are an abnormality detection device, an abnormality detection method, an abnormality detection system, and a storage medium. The abnormality detection device includes: a division unit that generates a plurality of period data in which normal vibration data for displaying normal vibration of the machine is divided into a plurality of periods; a transforming unit that performs fast Fourier transform on each of the plurality of period data generated by the dividing unit to calculate a plurality of power spectra for each period; a characteristic spectrum calculation unit that calculates one or more characteristic spectra for each period based on the plurality of power spectra calculated by the conversion unit; a model generation unit that generates a normal model for detecting an abnormality occurring in the machine from the one or more characteristic spectra calculated by the characteristic spectrum calculation unit; an index value calculation unit that calculates a predetermined index value from the normal model and vibration data for displaying vibration of the machine; and a determination unit configured to determine whether or not an abnormality has occurred in the machine, based on the index value and a predetermined threshold value set in advance.
Description
Technical Field
The present invention relates to an abnormality detection device, an abnormality detection method, an abnormality detection system, and a storage medium storing an abnormality detection program.
Background
For example, there is known a technique for detecting an abnormal vibration in a device such as a generator or a motor, thereby detecting an abnormality in the device such as a component life or a component deterioration (see, for example, patent documents 1 and 2). In such a technique, for example, occurrence of abnormal vibration can be detected by using a spectrum waveform of vibration occurring in the machine and depending on whether or not a spectrum of a frequency band for displaying abnormal vibration exceeds a threshold.
[ Prior art documents ]
[ patent document ]
[ patent document 1] (Japanese patent application laid-open No. 2009-128103)
[ patent document 2] (Japanese patent application laid-open No. 2011-22160)
Disclosure of Invention
[ problems to be solved by the invention ]
However, in the above-described conventional technique, the occurrence of abnormal vibration may not be detected with high accuracy. For example, in an industrial robot that performs a complicated operation such as a vertical articulated robot, vibration components (components) in the x-axis direction, the y-axis direction, and the z-axis direction may be related to each other. In such a case, even if the frequency spectrum of the frequency band indicating abnormal vibration in a certain vibration component exceeds the threshold, it cannot be said that an abnormality has occurred in the device.
In view of the above problems, an object of one embodiment of the present invention is to detect the occurrence of an abnormality with high accuracy from the vibration of the machine.
[ means for solving problems ]
In order to achieve the above object, one embodiment of the present invention is an abnormality detection device for performing abnormality detection based on vibration data for displaying vibration of an apparatus, including:
a dividing unit that generates a plurality of period data that divides normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width;
a transforming unit that performs a fast fourier transform using a window function on each of the plurality of period data generated by the dividing unit, and calculates a plurality of power spectra for each of the periods;
a characteristic spectrum calculation unit that calculates one or more characteristic spectra for each of the periods based on the plurality of power spectra calculated by the conversion unit;
a model generation unit that generates a normal model for detecting an abnormality occurring in the machine, based on the one or more characteristic spectra calculated by the characteristic spectrum calculation unit;
an index value calculation unit that calculates a predetermined index value based on the normal model generated by the model generation unit and vibration data for displaying vibration of the machine; and
and a determination unit configured to determine whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation unit and a predetermined threshold value set in advance.
[ Effect of the invention ]
The occurrence of an abnormality can be detected with high accuracy from the vibration of the machine.
Drawings
Fig. 1 is a diagram showing an example of the overall configuration of an abnormality detection system according to a first embodiment.
Fig. 2 is a diagram showing an example of a hardware configuration of the abnormality detection device according to the first embodiment.
Fig. 3 is a diagram showing an example of a functional configuration of the abnormality detection device according to the first embodiment.
FIG. 4 is a diagram showing an example of vibration data.
Fig. 5 is a flowchart showing an example of the model generation process according to the first embodiment.
FIG. 6 is a diagram showing an example of an average power spectrum.
FIG. 7 is a diagram showing an example of a maximum power spectrum.
Fig. 8 is a flowchart showing an example of the abnormality detection processing according to the first embodiment.
FIG. 9 is a diagram showing an example of a maximum Q value for each period.
FIG. 10 is a diagram showing an example of Q values for each frequency number in one period.
FIG. 11 is a diagram showing an example of a contribution diagram for each frequency number in one period.
Fig. 12 is a diagram showing another example of the overall configuration of the abnormality detection system according to the first embodiment.
Fig. 13 is a diagram showing an example of a functional configuration of an abnormality detection device according to a second embodiment.
Fig. 14 is a flowchart showing an example of the abnormality detection processing according to the second embodiment.
FIG. 15 shows an example of an output result.
FIG. 16 is a diagram showing another example of an output result.
[ description of symbols ]
1 abnormality detection system
10 abnormality detection device
20 perception machine (sending machine)
30 object machine
101 data obtaining part
102 frequency conversion unit
103 characteristic spectrum calculating part
104 model generation unit
105 index value calculation unit
106 abnormality determination unit
107 output unit
110 vibration data storage part
120 model storage unit
Detailed Description
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[ first embodiment ]
< overall construction >
First, the overall configuration of the abnormality detection system 1 according to the present embodiment will be described with reference to fig. 1. Fig. 1 is a diagram showing an example of the overall configuration of an abnormality detection system 1 according to a first embodiment.
As shown in fig. 1, an abnormality detection system 1 of the present embodiment includes an abnormality detection device 10 and a sensing machine 20. The abnormality detection device 10 and the sensing device 20 are communicably connected via a network such as a lan (local Area network).
The perception device 20 is a measuring device that measures vibration of the target device 30, which is an object to be detected whether or not an abnormality has occurred. The perception device 20 is, for example, a three-axis acceleration sensor or the like, and measures acceleration of the target device 30 in the x-axis direction, acceleration of the target device 30 in the y-axis direction, and acceleration of the target device 30 in the z-axis direction to generate vibration data including vibration data of the x-axis component, vibration data of the y-axis component, and vibration data of the z-axis component. Hereinafter, vibration data of an x-axis component is referred to as "x-component vibration data", vibration data of a y-axis component is referred to as "y-component vibration data", and vibration data of a z-axis component is referred to as "z-component vibration data".
Further, the perception appliance 20 may transmit the generated vibration data to the abnormality detection apparatus 10. It should be noted that the perception machine 20 may perform measurement of the object machine 30 at each predetermined time (i.e., each sampling period) determined in advance, for example, and generate vibration data.
The vibration data is not limited to the case including the three axes of acceleration. The vibration data may also include, for example, displacement (e.g., x-direction displacement, y-direction displacement, and z-direction displacement), and may also include velocity (e.g., x-direction velocity, y-direction velocity, and z-direction velocity).
The target device 30 is a device or equipment installed in a factory, a workshop, or the like. Specific examples of the target machine 30 include a working machine (e.g., a cutting machine, a bending machine, etc.), an industrial machine (e.g., a conveyor, a roller press, etc.), a semiconductor manufacturing apparatus, an electric heating apparatus, a production robot (e.g., a vertical articulated robot, a horizontal articulated robot, etc.), and the like. The target device 30 may be an inspection device that performs an inspection using vibration, or a device in a vehicle form such as a train vehicle.
The abnormality detection device 10 is a computer that detects an abnormality occurring in the target device 30 from vibration data received from the perception device 20. As the abnormality detection device 10, for example, a control device such as a plc (programmable Logic controller) may be used.
The operation of the present system includes: a "model generation" stage of generating a normal model for detecting an abnormality occurring in the target device 30; and an "evaluation" stage of detecting an abnormality based on the normal model and vibration data including acceleration measured during the operation of the target device 30 or the like. In general, the "model generation" stage is an off-line (offline) process performed when the target machine 30 is not operating, and the "evaluation" stage is an on-line (online) process performed during the operation of the target machine 30. However, the present invention is not limited to this, and both the "model generation" stage and the "evaluation" stage may be processed off-line, and both the "model generation" stage and the "evaluation" stage may be processed on-line.
The abnormality detection device 10 can generate a normal model from the vibration data for model generation in the "model generation" stage. The model generation vibration data is, for example, vibration data generated by measuring the target device 30 that is operating normally by the sensing device 20. The vibration data for model generation may be vibration data generated by a user or the like as data indicating normal operation of the target device 30.
In addition, the abnormality detection device 10 can detect an abnormality occurring in the target device 30 based on the evaluation vibration data and the normal model in the "evaluation" stage. The evaluation vibration data is, for example, vibration data generated by the sensing device 20 measuring the target device 30 during the online operation of the target device 30.
The abnormality detection system 1 of the present embodiment may include a plurality of (complex) types of target devices 30. In this case, the abnormality detection device 10 of the present embodiment may generate a normal model for each type of the target device 30 and detect an abnormality occurring in the target device 30 for each type of the target device 30.
In addition, the target device 30 may execute a plurality of types of operations. For example, the target machine 30 that manufactures a product by a plurality of steps may perform the operation a of step a, the operation B of step B, and the operation C of step C. In this case, the abnormality detection device 10 according to the present embodiment may generate a normal model for each operation and detect an abnormality occurring in the target device 30 for each operation.
In addition, the vibration of one object machine 30 may be measured by a plurality of perception machines 20. In this case, the abnormality detection device 10 may collect a plurality of vibration data generated by the plurality of sensing devices 20 into one data, and generate a normal model and/or detect an abnormality based on the collected data. The one piece of data after the collection is, for example, data including x-component vibration data, y-component vibration data, and z-component vibration data measured by the 1 st sensing machine 20 and x-component vibration data, y-component vibration data, and z-component vibration data measured by the 2 nd sensing machine 20 when vibration of one target machine 30 is measured by two sensing machines 20.
< hardware construction >
Next, the hardware configuration of the abnormality detection device 10 according to the present embodiment will be described with reference to fig. 2. Fig. 2 is a diagram showing an example of the hardware configuration of the abnormality detection device 10 according to the first embodiment.
As shown in fig. 2, the abnormality detection device 10 of the present embodiment includes an input device 11, a display device 12, an external I/F13, a ram (random Access memory)14, a rom (read Only memory)15, a cpu (central Processing unit)16, a communication I/F17, and an auxiliary storage device 18. These pieces of hardware are communicably connected by a bus 19.
The input device 11 is, for example, a keyboard, a mouse, a touch screen, or the like, and is used for a user to input various operations. The display device 12 is, for example, an lcd (liquid Crystal display) or the like, and displays a processing result of the abnormality detection device 10. The abnormality detection device 10 may not include at least one of the input device 11 and the display device 12.
The external I/F13 is an interface for interacting with external devices. The external device has a storage (recording) medium 13a and the like. The abnormality detection device 10 can read and write the storage medium 13a via the external I/F13. The storage medium 13a may be, for example, a flexible disk, a cd (compact disc), a dvd (digital Versatile disc), an SD memory card, a USB memory, or the like. The storage medium 13a may store a program for implementing various functions of the abnormality detection device 10 according to the present embodiment and/or a program for implementing the abnormality detection method according to the present embodiment.
The RAM14 is a volatile semiconductor memory for temporarily storing programs and data. The ROM15 is a nonvolatile semiconductor memory that can store programs and/or data even when power is turned off. The ROM15 stores, for example, BIOS (Basic Input/Output System), os (operating System) settings, network settings, and the like that are executed when the abnormality detection device 10 is activated.
The CPU16 is a computing device that reads out programs and/or data from the ROM15, the auxiliary storage device 18, and the like onto the RAM14 and executes processing, thereby realizing the overall control and/or function of the abnormality detection device 10.
The communication I/F17 is an interface for the abnormality detection device 10 to communicate with other devices and the like. The abnormality detection device 10 may receive vibration data from the perception appliance 20 via the communication I/F17.
The auxiliary storage device 18 is a nonvolatile memory in which programs and/or data are stored, and may be, for example, an hdd (hard Disk drive), an ssd (solid state drive), or the like. The programs and/or data stored in the auxiliary storage device 18 include programs for implementing various functions of the abnormality detection device 10 of the present embodiment, programs for implementing the abnormality detection method of the present embodiment, an OS that is basic software for overall control of the abnormality detection device 10, application software for providing various functions on the OS, and the like. The auxiliary storage device 18 may manage the stored programs and/or data by a predetermined file system, a DB (database), or the like.
The abnormality detection device 10 of the present embodiment has the above-described hardware configuration, and can realize various processes as described below.
< functional constitution >
Next, the functional configuration of the abnormality detection device 10 according to the present embodiment will be described with reference to fig. 3. Fig. 3 is a diagram showing an example of a functional configuration of the abnormality detection device 10 according to the first embodiment.
As shown in fig. 3, the abnormality detection device 10 of the present embodiment includes a data acquisition unit 101, a frequency conversion unit 102, a characteristic spectrum calculation unit 103, a model generation unit 104, an index value calculation unit 105, an abnormality determination unit 106, and an output unit 107. Each of these functional units can be realized by a process of causing the CPU16 to execute one or more programs installed in the abnormality detection device 10.
The abnormality detection device 10 of the present embodiment further includes a vibration data storage unit 110 and a model storage unit 120. These storage sections can be realized by using the auxiliary storage device 18, for example. At least one of these storage units may be realized by using a storage device or the like connected to the abnormality detection device 10 via a network.
The vibration data storage unit 110 stores the model generation vibration data and the evaluation vibration data. These model generation vibration data and evaluation vibration data are time-series data including accelerations (acceleration in the x-axis direction, acceleration in the y-axis direction, and acceleration in the axial direction) measured by the sensing machine 20 for each sampling period. In other words, the vibration data can be represented in a time domain in which the horizontal axis represents time and the vertical axis represents acceleration (acceleration in the x-axis direction, acceleration in the y-axis direction, or acceleration in the z-axis direction).
The vibration data for model generation and the vibration data for evaluation are not required to be stored in the vibration data storage unit 110 as separate data. For example, in one vibration data, data of a certain time width (for example, including time t ═ t) may be used1To time t ═ t2Vibration data of acceleration measured during the period of (a)) as vibration data for model generation, and data of another time width (including, for example, time t-t)3To time t ═ t4The vibration data of the acceleration measured during the period) as the evaluation vibration data.
Fig. 4 shows an example of vibration data (vibration data for model generation or vibration data for evaluation) stored in the vibration data storage unit 110. Fig. 4 is a diagram showing an example of vibration data.
Fig. 4(a) shows an example of x-component vibration data included in the vibration data. Fig. 4(b) shows an example of y-component vibration data included in the vibration data. Fig. 4(c) shows an example of z-component vibration data included in the vibration data. As shown in fig. 4(a) to 4(b), the vibration data of each component is time-series data in which the horizontal axis represents time and the vertical axis represents acceleration of each component. The time width of one vibration data (i.e., the time width from the measurement start time to the measurement end time of the acceleration included in the vibration data) is represented as a sampling period.
The data obtaining unit 101 can obtain the vibration data for model generation from the vibration data storage unit 110 at the "model generation" stage. The data obtaining unit 101 can obtain the vibration data for evaluation from the vibration data storage unit 110 at the "evaluation" stage.
The frequency conversion unit 102 generates period data obtained by dividing the sampling period of the vibration data (vibration data for model generation or vibration data for evaluation) obtained by the data obtaining unit 101 into predetermined period units in the "model generation" stage and the "evaluation" stage. The frequency Transform unit 102 performs Fast Fourier Transform (FFT) using a window function on the period-by-period data, and transforms the period data into the frequency domain for each window.
Accordingly, a power spectrum with a spectrum intensity on the vertical axis and a frequency on the horizontal axis can be obtained for each window in one period data. For example, in the case where L windows are included in one period data, L power spectra can be obtained.
It should be noted that the period data may be generated in accordance with the vibration data of each component. For example, when the sampling period of the vibration data obtained by the data obtaining unit 101 is divided into N periods of data, the sampling period of the x-component vibration data, the sampling period of the y-component vibration data, and the sampling period of the z-component vibration data may be divided into N periods to generate the period data. Therefore, in this case, N period data obtained by N-dividing the x-component vibration data, N period data obtained by N-dividing the y-component vibration data, and N period data obtained by N-dividing the z-component vibration data can be generated.
The characteristic spectrum calculation unit 103 calculates a characteristic power spectrum indicating a predetermined characteristic from the power spectrum obtained by the frequency conversion unit 102 for each period data in the "model generation" stage and the "evaluation" stage. As the characteristic power spectrum, an average power spectrum indicating an average value of the power spectra obtained for each window and a maximum power spectrum indicating a maximum value of the power spectra obtained for each window can be cited. This makes it possible to obtain a characteristic power spectrum for each period divided by the frequency converter 102.
The model generation unit 104 generates a normal model from the characteristic spectrum (a plurality of characteristic spectra in a period unit) obtained by the characteristic spectrum calculation unit 103 at the "model generation" stage. In this case, the model generation unit 104 may generate a normal model from a plurality of characteristic spectra using, for example, the method disclosed in japanese patent application laid-open No. 2016-.
After that, the model generation unit 104 stores the generated normal model in the model storage unit 120. The normal model is also referred to as "Profile".
The index value calculation unit 105 calculates a predetermined index value for each period based on the normal model stored in the model storage unit 120 and the characteristic power spectrum obtained by the characteristic spectrum calculation unit 103 at the "evaluation" stage. The predetermined index value may be a Q statistic (Q value) for each frequency band in the period and/or a maximum value (Q maximum value) of Q values in the period.
The abnormality determination section 106 may determine whether the index value calculated by the index value calculation section 105 exceeds a predetermined threshold value. When it is determined that the index value exceeds the threshold value, it is possible to detect that an abnormality has occurred in the target device 30.
The output unit 107 can output, for example, a graph or the like in which the index value calculated by the index value calculation unit 105 is plotted. Examples of the output target include the display device 12 and the like.
< model creation processing >
Next, a model generation process for generating a normal model will be described with reference to fig. 5. Fig. 5 is a flowchart showing an example of the model generation processing according to the first embodiment.
First, the data obtaining unit 101 obtains vibration data for model generation from the vibration data storage unit 110 (step S101). As described above, the vibration data for model generation includes x-component vibration data, y-component vibration data, and z-component vibration data.
Then, the frequency conversion unit 102 generates period data in which the sampling period of the vibration data (vibration data for model generation) obtained by the data obtaining unit 101 is divided into predetermined period units (step S102). Here, the predetermined period may be a time width including 65536 data values (acceleration values), for example. Next, a time width including 65536 data values is defined as one period.
For example, when the sampling period is divided into N periods of period 1 to period N, N period data obtained by N-dividing the x-component vibration data, N period data obtained by N-dividing the y-component vibration data, and N period data obtained by N-dividing the z-component vibration data can be generated. In this embodiment, a case where a sampling period is divided into N periods of 1 to N is described as a column.
Hereinafter, each period data divided based on the x-component vibration data is represented as "x-component period data", each period data divided based on the y-component vibration data is represented as "y-component period data", and each period data divided based on the z-component vibration data is represented as "z-component period data".
Next, the frequency converter 102 performs fast fourier transform using a window function for each period data, thereby calculating a power spectrum converted into a frequency domain for each window (step S103).
More specifically, the frequency converter 102 performs fast fourier transform for each window to calculate a power spectrum, for example, by setting a time width including 2048 data values as a window width and setting an overlap (overlap) rate to 50%. From this, for example, for one period data, L65536/(2048/2) 64 power spectra can be calculated. That is, L power spectra may be calculated for each of the N x-component period data. Likewise, L power spectra may be calculated for each of the N y-component period data. Likewise, L power spectra may be calculated for each of the N z-component period data.
Then, the characteristic spectrum calculation unit 103 calculates a characteristic power spectrum indicating a predetermined characteristic from the power spectrum calculated by the frequency conversion unit 102 for each period data (step S104). Next, as the characteristic power spectrum, an average power spectrum and a maximum power spectrum are calculated.
The average power spectrum can be obtained by averaging the L power spectra according to the data of each period. More specifically, with respect to L power spectra calculated based on one period data, an average power spectrum can be obtained by calculating an average of the spectral intensities for each frequency.
The maximum power spectrum can be obtained by calculating the maximum value of the L power spectra in accordance with the data of each period. More specifically, with respect to L power spectra calculated based on one period data, a maximum power spectrum can be obtained by calculating the maximum value of the spectral intensity by each frequency.
From this, an average power spectrum and a maximum power spectrum can be calculated for each of the N x-component period data. Likewise, an average power spectrum and a maximum power spectrum may be calculated for each of the N y-component period data. Likewise, an average power spectrum and a maximum power spectrum may be calculated for each of the data during the N z-components.
Note that the characteristic spectrum calculation unit 103 may calculate either one of the average power spectrum and the maximum power spectrum. This is because, for example, when it is known in advance that abnormal vibration occurring in the target device 30 is steady vibration, abnormality of the target device 30 can be detected by using a normal model generated based on the average power spectrum. Similarly, for example, when it is previously known that the abnormal vibration occurring in the target device 30 is a sudden vibration, the abnormality of the target device 30 can be detected by using a normal model generated based on the maximum power spectrum.
Here, an example of the average power spectrum is shown in fig. 6. Fig. 6 is a diagram showing an example of the average power spectrum.
Fig. 6(a) is an average power spectrum calculated from data during one x component. Fig. 6(b) is an average power spectrum calculated from data during one y-component. Fig. 6(c) is an average power spectrum calculated from data during one z-component. As shown in fig. 6(a) to 6(c), the average power spectrum of each component is data in which the horizontal axis represents the frequency number and the vertical axis represents the spectrum intensity. The frequency number is a number indicating a predetermined frequency band.
Hereinafter, an average power spectrum calculated from data during one x-component period is referred to as "x-component average power spectrum", an average power spectrum calculated from data during one y-component period is referred to as "y-component average power spectrum", and an average power spectrum calculated from data during one z-component period is referred to as "z-component average power spectrum".
Fig. 7 shows an example of the maximum power spectrum. Fig. 7 is a diagram showing an example of the maximum power spectrum.
Fig. 7(a) is a maximum power spectrum calculated from data during one x component. Fig. 7(b) is a maximum power spectrum calculated from data during one y component. Fig. 7(c) is a maximum power spectrum calculated from data during one z-component. As shown in fig. 7(a) to 7(c), the maximum power spectrum of each component is data in which the horizontal axis represents a frequency number and the vertical axis represents a spectrum intensity.
Hereinafter, a maximum power spectrum calculated from data during one x-component period is referred to as "x-component maximum power spectrum", a maximum power spectrum calculated from data during one y-component period is referred to as "y-component maximum power spectrum", and a maximum power spectrum calculated from data during one z-component period is referred to as "z-component maximum power spectrum".
As shown in fig. 6 and 7, the maximum power spectrum has a higher spectrum intensity at each frequency number than the average power spectrum.
Note that the characteristic spectrum calculation unit 103 may calculate, as the characteristic spectrum, for example, a standard deviation power spectrum, a maximum variation ratio power spectrum, or the like, in addition to the average power spectrum and the maximum power spectrum. The standard deviation power spectrum refers to a power spectrum obtained by calculating a standard deviation of the spectral intensity for each frequency for L power spectra calculated based on one period data. The maximum variation ratio power spectrum is a power spectrum obtained by calculating the maximum value of the difference in spectral intensity for each frequency between adjacent windows (windows that are partially overlapped).
Next, the model generation unit 104 generates a normal model from the plurality of average power spectra and the plurality of maximum power spectra calculated by the characteristic spectrum calculation unit 103 (step S105). At this time, the model generation unit 104 may generate a normal model from a plurality of average power spectra and a plurality of maximum power spectra by using, for example, the method disclosed in japanese patent application laid-open No. 2016-.
For example, the N x-component average power spectra, the N y-component average power spectra, the N z-component average power spectra, the N x-component maximum power spectra, the N y-component maximum power spectra, and the N z-component maximum power spectra may be obtained by using the model generation method disclosed in japanese patent application laid-open No. 2016-164772 (japan) as six-variable batch data representing N batches (batch).
More specifically, when N is 1, ·, and N, the x-component average power spectrum and the x-component maximum power spectrum obtained from the x-component period data of the period N, the y-component average power spectrum and the y-component maximum power spectrum obtained from the y-component period data of the period N, and the z-component average power spectrum and the z-component maximum power spectrum obtained from the z-component period data of the period N may be set as six-variable batch data of one batch within the period N, and the model generation method disclosed in (japanese) japanese patent application laid-open No. 2016-.
Accordingly, the normal model can be generated by the model generation unit 104. The generated normal model is stored in the model storage unit 120.
In the present embodiment, the case where two power spectra, i.e., the average power spectrum and the maximum power spectrum, are used as the characteristic power spectrum is described, but, for example, in the case where only one of the average power spectrum and the maximum power spectrum is used, the model generation method disclosed in japanese patent application laid-open No. 2016-164772 (japanese) may be adopted as the three-variable batch data representing N batches. The number (number) of variables included in one batch is determined by the number of characteristic power spectra and the number of variables included in the vibration data. For example, when the number of characteristic power spectra is S and the number of variables included in the vibration data is T, the number of variables included in one batch is S × T.
< abnormality detection processing >
Next, an abnormality detection process for detecting occurrence of an abnormality in the target device 30 using the normal model will be described with reference to fig. 8. Fig. 8 is a flowchart showing an example of the abnormality detection processing according to the first embodiment.
First, the data obtaining unit 101 obtains vibration data for evaluation from the vibration data storage unit 110 (step S201). As described above, the evaluation vibration data includes x-component vibration data, y-component vibration data, and z-component vibration data.
Then, the frequency conversion unit 102 generates period data in which the sampling period of the vibration data (evaluation vibration data) obtained by the data obtaining unit 101 is divided into predetermined period units (step S202). The predetermined period is, for example, a time width including 65536 data values (acceleration values) as in the "model generation" stage.
Hereinafter, as period data of M periods in which the sampling period is divided into the period 1 to the period M, similarly to the "model generation" stage, the period data obtained by dividing the x-component vibration data is represented as "x-component period data", the period data obtained by dividing the y-component vibration data is represented as "y-component period data", and the period data obtained by dividing the z-component vibration data is represented as "z-component period data".
Next, the frequency transform unit 102 performs fast fourier transform using a window function on each period data, thereby calculating a power spectrum transformed into a frequency domain for each window (step S203). In this case, the frequency transform unit 102 performs fast fourier transform using the same window width and overlap ratio as in the "model generation" stage. Accordingly, as in the "model generation" stage, for example, L power spectra can be calculated for one period of data.
In the present embodiment, a case where fast fourier transform using a window function is performed on data of each period is described, but the present invention is not limited to this case. For example, wavelet transform (wavelet transform) or the like may be performed for each period data.
Then, the characteristic spectrum calculation unit 103 calculates a characteristic power spectrum indicating a predetermined characteristic from the power spectrum calculated by the frequency conversion unit 102 for each period data (step S204). In this case, the characteristic spectrum calculating unit 103 may calculate a characteristic power spectrum similar to the "model generation" stage. Next, as the characteristic power spectrum, an average power spectrum and a maximum power spectrum are calculated.
Accordingly, an average power spectrum and a maximum power spectrum may be calculated for each of the M x-component period data. Likewise, an average power spectrum and a maximum power spectrum may be calculated for each of the M y-component period data. Likewise, an average power spectrum and a maximum power spectrum may be calculated for each of the M z-component period data.
Next, the index value calculation unit 105 calculates a predetermined index value for each period based on the normal model stored in the model storage unit 120 and the characteristic power spectrum (average power spectrum and maximum power spectrum) obtained by the characteristic spectrum calculation unit 103 (step S205). Next, a case of calculating, as the predetermined index value, a Q value for each frequency number indicating the frequency band in the period and a Q maximum value for each period number indicating the period will be described. However, the index value is not limited to the Q statistic and/or the maximum value of the Q statistic, and for example, T may be used2Statistic, T2The maximum value of the statistic, etc.
The Q value of each frequency number in a certain period can be represented by the contribution diagram of each frequency number of each characteristic power spectrum(distribution plot) (contribution plot of Q statistic) is shown in total. For example, if the contribution of a certain frequency number f in the average power spectrum of the x component is "Q" in this period11(f) The contribution of the frequency number f in the average power spectrum of the y-component is "" Q ""12(f) The contribution of the frequency number f in the average power spectrum of the z-component is "" Q ""13(f) The contribution of the frequency number f in the maximum power spectrum of the x-component is "" Q ""21(f) The contribution of the frequency number f in the maximum power spectrum of the y-component is "" Q ""22(f) And the contribution of the frequency number f in the maximum power spectrum of the z-component is "Q23(f) Then the Q value of the frequency number f can be selected from Q11(f)+Q12(f)+Q13(f)+Q21(f)+Q22(f)+Q23(f) And (4) showing.
The maximum Q value for each period number is the maximum value of Q (f) in the period n indicated by the period number n.
Then, the abnormality determination unit 106 determines whether the index value calculated by the index value calculation unit 105 exceeds a predetermined threshold value (step S206). The threshold value may be set for each index value. That is, when the Q value and the Q maximum value are used as the index values, a threshold value of the Q value for each frequency number and a threshold value of the Q maximum value for each period number may be set.
Next, the output unit 107 may output, for example, a graph in which the index value calculated by the index value calculation unit 105 is plotted (step S207).
Here, as an example of the output result of the output unit 107, a graph showing the maximum Q value for each period number is shown in fig. 9. The graph shown in fig. 9 is a graph in which the period number is plotted on the horizontal axis and the maximum Q value is plotted on the vertical axis. In the example shown in fig. 9, "10000" is set as a threshold value of the maximum value of Q for each period number. In this case, the abnormality determination unit 106 can detect that an abnormality has occurred at the period number "164" exceeding the threshold.
Accordingly, the user of the abnormality detection apparatus 10 of the present embodiment can know the period in which the abnormality has occurred in the target device 30.
Fig. 10 shows graphs showing Q values for each frequency number in a certain period as another example of the output result of the output unit 107. The graph shown in fig. 10 is a graph with the frequency number on the horizontal axis and the Q value on the vertical axis. In the example shown in fig. 10, "200" is set as the threshold value of the Q value for each frequency number. In this case, the abnormality determination unit 106 can detect that an abnormality has occurred at the frequency number "150" exceeding the threshold.
Accordingly, for example, when the type of abnormality and the frequency band in which the abnormality occurs are associated in advance, the user of the abnormality detection device 10 of the present embodiment can know the type of abnormality occurring in the target device 30. That is, when the frequency band is determined when a certain abnormality occurs in the target device 30, the user of the abnormality detection device 10 according to the present embodiment can also know the type of the abnormality occurring in the target device 30.
Here, as described above, the Q value of each frequency number may pass through the contribution graph Q at that frequency number11～Q13And Q21～Q23Represents the total of (a). An example of each contribution graph is shown in fig. 11. FIG. 11(a) is a contribution graph Q of each frequency number in a certain period11. FIG. 11(b) is a contribution chart Q of each frequency number in the period12. FIG. 11(c) is a contribution chart Q of each frequency number in the period13. FIG. 11(d) is a contribution chart Q of each frequency number in the period21. FIG. 11(e) is a contribution chart Q of each frequency number in the period22. FIG. 11(f) is a contribution chart Q of each frequency number in the period23. By calculating Q according to each frequency number11～Q13And Q21～Q23The Q value at that frequency number can be calculated. When the occurrence of an abnormality is detected, the user of the abnormality detection apparatus 10 according to the present embodiment can know which of the variables of which characteristic power spectrum has a high contribution degree by referring to the contribution map for each frequency number.
< other example of the abnormality detection system 1 >
Here, another example of the entire configuration of the abnormality detection system 1 according to the present embodiment will be described with reference to fig. 12. Fig. 12 is a diagram showing another example of the entire configuration of the abnormality detection system 1 according to the first embodiment.
As shown in fig. 12, the abnormality detection system 1 of the present embodiment includes an abnormality detection device 10, a sensing device 20, and a display device 40, and may be configured to be communicably connected via a network N such as the world wide web (Internet), for example. In other words, the model generation processing and the abnormality detection processing based on the abnormality detection apparatus 10 may be provided as a cloud serVice (cloud serVice) to the user of the display apparatus 40.
In the abnormality detection system 1 shown in fig. 12, the abnormality detection device 10 transmits the index value calculated by the index value calculation unit 105 and the determination result of the abnormality determination unit 106 to the display device 40. Accordingly, the display device 40 can display the output results shown by the graphs shown in fig. 9 to 11, for example. As the display device 40, for example, a PC (personal computer), a smart phone, a tablet terminal, or the like can be used.
< summary of the first embodiment >
As described above, the abnormality detection system 1 according to the present embodiment can generate a normal model from vibration data indicating normal operation of the target device 30, for example, offline. In addition, the abnormality detection system 1 of the present embodiment can also detect an abnormality of the target machine 30 based on the vibration data obtained by measuring the motion of the target machine 30 that operates on-line by the perception machine 20 and the normal model. Accordingly, in the abnormality detection system 1 of the present embodiment, the occurrence of an abnormality can be detected with high accuracy from the vibration of the target device 30.
[ second embodiment ]
Next, a second embodiment will be explained. In the second embodiment, a case will be described in which, when an abnormality is detected, a variable in which the abnormality has occurred (for example, a variable indicating an x component, a y component, a z component, and the like of acceleration) is specified, and a power spectrum and a normal model of the specified variable are displayed in addition to the specified variable. Accordingly, for example, the user of the abnormality detection apparatus 10 can check the power spectrum and the normal model of the abnormality-occurring variable, and can use the checked power spectrum and the normal model as references for finding the cause of the abnormality occurrence, specifying the position of the abnormality, and the like.
In the second embodiment, the description will be given mainly of matters different from those of the first embodiment. The same components as those in the first embodiment are denoted by the same reference numerals, and descriptions thereof are omitted.
< functional constitution >
First, a functional configuration of the abnormality detection device 10 according to the present embodiment will be described with reference to fig. 13. Fig. 13 is a diagram showing an example of a functional configuration of the abnormality detection device 10 according to the second embodiment.
As shown in fig. 13, the abnormality detection device 10 of the present embodiment further includes a specification unit 108. This functional unit can be realized by the CPU16 executing one or more programs installed in the abnormality detection device 10.
The specifying unit 108 can specify a variable in which an abnormality has occurred (more specifically, a variable with a high possibility of occurrence of an abnormality) when the abnormality determination unit 106 detects occurrence of an abnormality.
Further, the output unit 107 of the present embodiment can output, as an output result, the power spectrum of the variable specified by the specifying unit 108 and the normal model of the variable. The output target includes, for example, the display device 12. Accordingly, the power spectrum of the variable in which the abnormality has occurred (more specifically, the variable in which the possibility of occurrence of the abnormality is high) and the normal model of the variable can be displayed.
< abnormality detection processing >
Next, with reference to fig. 14, an abnormality detection process of detecting occurrence of an abnormality in the target device 30 using a normal model and displaying a power spectrum of a variable in which the abnormality has occurred and the normal model of the variable when the occurrence of the abnormality is detected will be described. Fig. 14 is a flowchart showing an example of the abnormality detection processing according to the second embodiment. Since steps S201 to S206 in fig. 14 are the same as those in fig. 8, their description will be omitted.
When an abnormality is detected by the abnormality determination unit 106 in step S206 (that is, when it is determined that the index value of a certain period ID exceeds the threshold value), the specification unit 108 specifies a frequency number in which an abnormality has occurred in the period indicated by the period ID (more specifically, a frequency number in which the possibility of occurrence of an abnormality is high) (step S301). Here, for example, when the index value is the Q maximum value, the determination unit 108 may determine, as the frequency number in which the abnormality has occurred, the frequency number having the highest Q value among the frequency numbers of the period ID. Similarly, for example, when the index value is a Q value, the determination unit 108 may determine a frequency number having the highest Q value among the frequency numbers of the period ID as the frequency number having the abnormality.
Then, the specification unit 108 specifies the variable having the highest degree of contribution to the Q value of the frequency number specified in step S301 in the period indicated by the period ID (step S302). If the frequency number is f, the Q value of the frequency number f can be represented by Q11(f)+Q12(f)+Q13(f)+Q21(f)+Q22(f)+Q23(f) And (4) showing. Therefore, the contribution of the variable x can be represented by Q11(f)+Q21(f) The degree of contribution of the variable y can be represented by Q12(f)+Q22(f) The contribution of the variable z can be represented by Q13(f)+Q23(f) And (4) showing. The determination unit 108 determines the variable having the highest contribution degree among the contribution degrees. Hereinafter, the variable specified by the specifying unit 108 is also referred to as "abnormality occurrence variable".
Next, the output section 107 outputs the power spectrum of the abnormality occurrence variable determined by the determination section 108 and the normal model of the abnormality occurrence variable as output results (step S303). Fig. 15 shows an example of the output result. As shown in fig. 15, as the output result, the power spectrum of the abnormality occurrence variable and the normal model of the abnormality occurrence variable are displayed in an overlapping manner. Accordingly, the user can compare the power spectrum of the abnormality occurrence variable with the normal model of the abnormality occurrence variable, and can use the power spectrum as a reference for finding the cause of the abnormality occurrence, specifying the position of the abnormality, and the like. In other words, the user can compare the behavior at the time of abnormality and the behavior at the time of normal in the frequency domain, thereby accurately confirming the actual abnormal situation.
In this case, the user may specify a range to be confirmed by, for example, a mouse or the like, and enlarge the specified range. Accordingly, the difference between the power spectrum of the abnormality occurrence variable and the normal model of the abnormality occurrence variable can be checked in more detail.
In this case, the user can also display the output results shown in fig. 16(a) to 16(c) by, for example, performing a display switching operation. The output result shown in fig. 16(a) is the power spectrum of the abnormality occurrence variable and the difference absolute value of each frequency of the normal model of the abnormality occurrence variable. The output result shown in fig. 16(b) is the ratio of the power spectrum of the abnormality occurrence variable to the frequencies of the normal model of the abnormality occurrence variable. The output result shown in fig. 16(c) is the difference between the power spectrum of the abnormality occurrence variable and each frequency of the normal model of the abnormality occurrence variable. By referring to these output results, the user can use them as a reference for finding the cause of the occurrence of the abnormality, specifying the position of the abnormality, and the like.
< summary of the second embodiment >
As described above, the abnormality detection system 1 according to the present embodiment can display the power spectrum of the abnormality occurrence variable and the normal model of the abnormality occurrence variable as the output result when an abnormality is detected. Accordingly, the user can compare the behavior at the time of abnormality in the frequency band with the behavior at the time of normal operation, and can use the comparison as a reference for finding the cause of the occurrence of abnormality, specifying the position of the abnormality, and the like.
In the present embodiment, while the abnormality has occurred, the variable that has the highest degree of contribution to the Q value of the frequency number specified in step S301 is used as the abnormality occurrence variable, but the present invention is not limited to this. For example, S variables located at the front (or rear) (i.e., having a high degree of contribution) may be used as the abnormality occurrence variables in the order of the degree of contribution to the Q value from high to low (or from low to high). Accordingly, for example, by displaying the power spectrum and the normal model of each abnormality occurrence variable in the order of the contribution degree to the Q value from high to low (or from low to high), the user can use them as a reference for finding the cause of the abnormality occurrence, specifying the position of the abnormality, and the like.
In view of the above, there is provided an abnormality detection device for performing abnormality detection based on vibration data indicating vibration of an apparatus, the abnormality detection device including: a dividing unit that generates a plurality of period data that divides normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width; a transforming unit that performs fast fourier transform using a window function on each of the plurality of period data generated by the dividing unit to calculate a plurality of power spectra for each of the periods; a characteristic spectrum calculation unit that calculates one or more characteristic spectra for each of the periods based on the plurality of power spectra calculated by the conversion unit; a model generation unit that generates a normal model for detecting an abnormality occurring in the machine, based on the one or more characteristic spectra calculated by the characteristic spectrum calculation unit; an index value calculation unit that calculates a predetermined index value based on the normal model generated by the model generation unit and vibration data for displaying vibration of the machine; and a determination unit configured to determine whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation unit and a predetermined threshold value set in advance.
The characteristic spectrum calculation unit calculates an average power spectrum indicating an average value of the plurality of power spectra and a maximum power spectrum indicating a maximum value of the plurality of power spectra for each of the periods based on the plurality of power spectra, and the index value calculation unit calculates a predetermined index value based on the average power spectrum and the maximum power spectrum calculated by the characteristic spectrum calculation unit and the normal model.
The index value calculation means calculates the Q statistic as the index value, the maximum value of the Q statistic in the period, and T2Statistics and T in the period2At least one of the maximum values of the statistics.
The abnormality detection apparatus further has an output unit for outputting at least one of the Q statistic for each frequency of the vibration and a maximum value of the Q statistic for each period.
The index value is a Q statistic or a maximum value of the Q statistic, the abnormality detection device further includes a specifying unit that specifies a frequency at which the Q statistic is maximum during a period in which the abnormality occurs, and specifies a predetermined number of variables in order of a high degree of contribution of the Q statistic to the specified frequency, when the determining unit determines that the abnormality has occurred in the machine, and the output unit outputs a power spectrum corresponding to the variable specified by the specifying unit and a normal model corresponding to the variable specified by the specifying unit among the plurality of power spectra calculated by the converting unit.
The output unit outputs at least one of a power spectrum corresponding to the variable and a difference absolute value, a ratio, and a difference of each frequency of the normal model corresponding to the variable.
There is also provided an abnormality detection method in which an abnormality detection device that detects an abnormality based on vibration data indicating vibration of a machine performs the steps of: a dividing step of generating a plurality of period data for dividing normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width; a transform step of performing fast fourier transform using a window function on each of the plurality of period data generated by the division step to calculate a plurality of power spectra for each of the periods; a characteristic spectrum calculation step of calculating one or more characteristic spectra for each period based on the plurality of power spectra calculated by the conversion step; a model generation step of generating a normal model for detecting an abnormality occurring in the machine from the one or more characteristic spectra calculated in the characteristic spectrum calculation step; an index value calculation step of calculating a predetermined index value from the normal model generated by the model generation step and vibration data for displaying vibration of the machine; and a determination step of determining whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation step and a predetermined threshold value set in advance.
There is also provided an abnormality detection system including a machine and a measuring machine that measures vibration of the machine, having: a dividing unit that generates a plurality of period data that divides normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width; a transforming unit that performs fast fourier transform using a window function on each of the plurality of period data generated by the dividing unit to calculate a plurality of power spectra for each of the periods; a characteristic spectrum calculation unit that calculates one or more characteristic spectra for each of the periods based on the plurality of power spectra calculated by the conversion unit; a model generation unit that generates a normal model for detecting an abnormality occurring in the machine, based on the one or more characteristic spectra calculated by the characteristic spectrum calculation unit; an index value calculation unit that calculates a predetermined index value based on the normal model generated by the model generation unit and vibration data for displaying vibration of the machine; and a determination unit configured to determine whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation unit and a predetermined threshold value set in advance.
Further, there is provided a storage medium storing an abnormality detection program, the abnormality detection program being a computer-readable program for causing a computer to execute the abnormality detection method.
Although the present invention has been described in connection with the above-described exemplary embodiments, the present invention is not limited to the specifically disclosed embodiments, and various modifications and/or changes may be made thereto without departing from the scope of the claims.
Claims (9)
1. An abnormality detection device that performs abnormality detection based on vibration data for displaying vibration of a machine, the abnormality detection device comprising:
a dividing unit that generates a plurality of period data that divides normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width;
a transforming unit that performs fast fourier transform using a window function on each of the plurality of period data generated by the dividing unit to calculate a plurality of power spectra for each of the periods;
a characteristic spectrum calculation unit that calculates one or more characteristic power spectra for each of the periods based on the plurality of power spectra calculated by the conversion unit;
a model generation unit that generates a normal model for detecting an abnormality occurring in the machine, based on the one or more characteristic power spectrums calculated by the characteristic spectrum calculation unit;
an index value calculation unit that calculates a predetermined index value based on the normal model generated by the model generation unit and vibration data for displaying vibration of the machine; and
and a determination unit configured to determine whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation unit and a predetermined threshold value set in advance.
2. The abnormality detection device according to claim 1,
the characteristic spectrum calculation unit calculates an average power spectrum indicating an average value of the plurality of power spectra and a maximum power spectrum indicating a maximum value of the plurality of power spectra for each of the periods based on the plurality of power spectra,
the index value calculation unit calculates a predetermined index value based on the average power spectrum and the maximum power spectrum calculated by the characteristic spectrum calculation unit and the normal model.
3. The abnormality detection device according to claim 1 or 2,
the index value calculation unit calculates a Q statistic, a maximum value of the Q statistic in the period, and T2Statistics and T in the period2Among the maximum values of the statisticsAs the index value.
4. The abnormality detection device according to claim 3, further comprising:
an output unit that outputs at least one of the Q statistic for each frequency of the vibration and a maximum value of the Q statistic for each period.
5. The abnormality detection device according to claim 4,
the index value is a Q statistic or a maximum value of the Q statistic,
the abnormality detection device further includes a specifying unit that specifies a frequency at which the Q statistic is the largest during a period in which the abnormality occurs, and specifies a predetermined number of variables in descending order of the degree of contribution to the Q statistic with respect to the specified frequency, when the determining unit determines that the abnormality has occurred in the machine,
the output unit outputs a power spectrum corresponding to the variable determined by the determination unit and a normal model corresponding to the variable determined by the determination unit among the plurality of power spectra calculated by the transformation unit.
6. The abnormality detection device according to claim 5,
the output unit outputs at least one of a power spectrum corresponding to the variable and a difference absolute value, a ratio, and a difference of each frequency of the normal model corresponding to the variable.
7. An abnormality detection method in which an abnormality detection device that performs abnormality detection based on vibration data for displaying vibration of a machine performs the steps of:
a dividing step of generating a plurality of period data for dividing normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width;
a transforming step of performing fast fourier transform using a window function on each of the plurality of period data generated by the dividing step to calculate a plurality of power spectra for each of the periods;
a characteristic spectrum calculation step of calculating one or more characteristic power spectra for each of the periods based on the plurality of power spectra calculated by the conversion step;
a model generation step of generating a normal model for detecting an abnormality occurring in the machine, based on one or more characteristic power spectrums calculated in the characteristic spectrum calculation step;
an index value calculation step of calculating a predetermined index value from the normal model generated by the model generation step and vibration data for displaying vibration of the machine; and
and a determination step of determining whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation step and a predetermined threshold value set in advance.
8. A storage medium storing a computer-readable program, wherein the computer-readable program causes a computer to execute the steps of:
a dividing step of generating a plurality of period data for dividing normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width;
a transforming step of performing fast fourier transform using a window function on each of the plurality of period data generated by the dividing step to calculate a plurality of power spectra for each of the periods;
a characteristic spectrum calculation step of calculating one or more characteristic power spectra for each of the periods based on the plurality of power spectra calculated by the conversion step;
a model generation step of generating a normal model for detecting an abnormality occurring in the machine, based on one or more characteristic power spectrums calculated in the characteristic spectrum calculation step;
an index value calculation step of calculating a predetermined index value from the normal model generated by the model generation step and vibration data for displaying vibration of the machine; and
and a determination step of determining whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation step and a predetermined threshold value set in advance.
9. An abnormality detection system including a machine and a measuring machine that measures vibration of the machine, the abnormality detection system having:
a dividing unit that generates a plurality of period data that divides normal vibration data for displaying normal vibration of the machine into a plurality of periods, each of the periods having a predetermined time width;
a transforming unit that performs fast fourier transform using a window function on each of the plurality of period data generated by the dividing unit to calculate a plurality of power spectra for each of the periods;
a characteristic spectrum calculation unit that calculates one or more characteristic power spectra for each of the periods based on the plurality of power spectra calculated by the conversion unit;
a model generation unit that generates a normal model for detecting an abnormality occurring in the machine, based on the one or more characteristic power spectrums calculated by the characteristic spectrum calculation unit;
an index value calculation unit that calculates a predetermined index value based on the normal model generated by the model generation unit and vibration data for displaying vibration of the machine; and
and a determination unit configured to determine whether or not an abnormality has occurred in the machine, based on the index value calculated by the index value calculation unit and a predetermined threshold value set in advance.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2017-221642 | 2017-11-17 | ||
| JP2017221642 | 2017-11-17 | ||
| JP2018-128340 | 2018-07-05 | ||
| JP2018128340A JP7358725B2 (en) | 2017-11-17 | 2018-07-05 | Abnormality detection device, abnormality detection method, abnormality detection program and abnormality detection system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109798970A CN109798970A (en) | 2019-05-24 |
| CN109798970B true CN109798970B (en) | 2021-04-02 |
Family
ID=66556234
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811248265.8A Active CN109798970B (en) | 2017-11-17 | 2018-10-24 | Abnormality detection device, abnormality detection method, abnormality detection system, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109798970B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112665706B (en) * | 2020-11-30 | 2023-04-11 | 武汉第二船舶设计研究所(中国船舶重工集团公司第七一九研究所) | Vibration monitoring and analyzing method and system for maritime work platform |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5862528A (en) * | 1981-10-09 | 1983-04-14 | Sumitomo Metal Ind Ltd | Monitoring method for periodical motion body |
| JP2004020424A (en) * | 2002-06-18 | 2004-01-22 | Mitsubishi Chemicals Corp | Processing method for vibration signal |
| CN102607845A (en) * | 2012-03-05 | 2012-07-25 | 北京工业大学 | Bearing fault characteristic extracting method for redundantly lifting wavelet transform based on self-adaptive fitting |
| CN102606891A (en) * | 2012-04-11 | 2012-07-25 | 广州东芝白云自动化系统有限公司 | Water leakage detector, water leakage detecting system and water leakage detecting method |
| CN104595112A (en) * | 2013-10-30 | 2015-05-06 | 通用电气公司 | Wind turbine and method for evaluating health status of blades thereon |
-
2018
- 2018-10-24 CN CN201811248265.8A patent/CN109798970B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5862528A (en) * | 1981-10-09 | 1983-04-14 | Sumitomo Metal Ind Ltd | Monitoring method for periodical motion body |
| JP2004020424A (en) * | 2002-06-18 | 2004-01-22 | Mitsubishi Chemicals Corp | Processing method for vibration signal |
| CN102607845A (en) * | 2012-03-05 | 2012-07-25 | 北京工业大学 | Bearing fault characteristic extracting method for redundantly lifting wavelet transform based on self-adaptive fitting |
| CN102606891A (en) * | 2012-04-11 | 2012-07-25 | 广州东芝白云自动化系统有限公司 | Water leakage detector, water leakage detecting system and water leakage detecting method |
| CN104595112A (en) * | 2013-10-30 | 2015-05-06 | 通用电气公司 | Wind turbine and method for evaluating health status of blades thereon |
Non-Patent Citations (1)
| Title |
|---|
| 发电机传动轴承的异常振动谱特征提取算法;陈运胜;《国外电子测量技术》;20160531;第35卷(第5期);第20-23、38页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109798970A (en) | 2019-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7358725B2 (en) | Abnormality detection device, abnormality detection method, abnormality detection program and abnormality detection system | |
| EP2836881B1 (en) | Embedded prognostics on plc platforms for equipment condition monitoring, diagnosis and time-to-failure/service prediction | |
| JP6371236B2 (en) | Predictive diagnostic system, predictive diagnostic method, and predictive diagnostic apparatus | |
| de Jesus et al. | Driver current analysis for sensorless tool breakage monitoring of CNC milling machines | |
| JP5991042B2 (en) | Anomaly monitoring system and anomaly monitoring method | |
| CN107614214B (en) | Trouble-shooter and method for diagnosing faults | |
| CN103901835A (en) | Apparatus and methods to monitor and control cyclic process units in a steady plant environment | |
| WO2019045699A1 (en) | Recurrent gaussian mixture model for sensor state estimation in condition monitoring | |
| CN109798970B (en) | Abnormality detection device, abnormality detection method, abnormality detection system, and storage medium | |
| JP7036261B1 (en) | Anomaly detection device, program, anomaly detection method, and manufacturing method | |
| US11941495B2 (en) | Information processing device, information processing method, and recording medium | |
| TW201633025A (en) | Diagnostic method for malfunction mode of machine tool main shaft and system thereof | |
| JP7230371B2 (en) | Abnormality detection device, abnormality detection method, abnormality detection program and abnormality detection system | |
| Shah et al. | 8 channel vibration monitoring and analyzing system using LabVIEW | |
| EP3819608A1 (en) | Detecting rotor anomalies by determining vibration trends during transient speed operation | |
| CN108595375A (en) | a kind of filtering method, device and storage medium | |
| US9829540B2 (en) | Method and controller for determining an undesired condition in an electrical drive system | |
| US20210349436A1 (en) | Sensor device, data acquisition method, and current monitoring system | |
| CN110703721B (en) | Method and device for monitoring a production process | |
| CN108169544B (en) | Voltage sag display method and device | |
| Xie et al. | Characteristic enhancement of RV reducer under non-stationary condition based on WDT-IF-SS-VMD | |
| US11487275B2 (en) | Analyzer | |
| TWI712944B (en) | Sound-based equipment surveillance method | |
| WO2022029873A1 (en) | Synchronizer, synchronization method and program therefor | |
| CN112836980A (en) | Data processing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |