System and method for detecting abnormal feature in rotating machine

Abstract

This disclosure provides a system and method for detecting abnormal features of rotating machinery. The system includes: a sensing module for acquiring vibration data of components of the rotating machinery; a data processing module for processing the vibration data to generate complex vibration time-series data; a conversion module for converting each of the complex vibration time-series data into complex converted data; and an abnormal feature detection module for comparing the feature value distribution in the complex converted data to define the component corresponding to the converted data where the feature value distribution changes as abnormal.

Description

Technical Field

[0001] The present invention discloses an abnormal characteristic detection system and method, and in particular, a rotating mechanical abnormal characteristic detection system and method. Prior Art

[0002] Components (such as gears, chains / belts, saw blades, drill bits, and tires) used in rotating machinery (e.g., power transmission equipment, conveyor belts, electric saws, electric drills, and transportation equipment) often become damaged after a period of operation, affecting the overall operation of the rotating machinery. Consequently, technologies have been developed to detect component damage. Currently, most existing technologies directly convert the vibration data detected from the components into frequency analysis. However, frequency analysis is not effective for rotating machinery operating at different speeds, as different speeds produce different frequencies, and there is no consistent basis for comparison. Furthermore, in electric saws, to prevent resonance during operation, the spacing between the multiple teeth on the saw blade must be uneven, making it difficult to determine the location of the damaged tooth based on the frequency. Summary of the Invention

[0003] The present disclosure provides a rotating machine abnormality feature detection system, comprising: a sensing module for acquiring vibration data of a component of the rotating machine; a data processing module for processing the vibration data to generate complex vibration time series data; a conversion module for converting each of the complex vibration time series data into complex conversion data; and an abnormality feature detection module for comparing the eigenvalue distribution in the complex conversion data to define the component corresponding to the conversion data having a change in the eigenvalue distribution as abnormal.

[0004] The present disclosure further provides a method for detecting abnormal characteristics of a rotating machine, comprising: causing a sensing module to obtain vibration data of a component of the rotating machine; causing a data processing module to process the vibration data to generate complex vibration time series data; causing a conversion module to convert each of the complex vibration time series data into complex conversion data; and causing an abnormal characteristic detection module to compare the characteristic value distribution in the complex conversion data to define the component corresponding to the conversion data in which the characteristic value distribution has changed as abnormal.

[0005] As described above, in the rotating machinery abnormal feature detection system and method, the abnormal feature detection module sequentially compares whether the characteristic value distribution of the same spacing interval in any two adjacent complex conversion data has changed, or first defines a complex conversion data group with N complex conversion data as a group, and then sequentially compares whether the characteristic value distribution of the same spacing interval in any two adjacent complex conversion data groups has changed, where N is a natural number.

[0006] As in the aforementioned rotating machinery abnormality characteristic detection system and method, the characteristic value distribution changes in that the complex third quartiles corresponding to the same interval in the complex conversion data or the complex conversion data group are different from each other.

[0007] As in the aforementioned rotating machinery abnormality feature detection system and method, the complex conversion data between the complex conversion data groups partially overlap or do not overlap at all.

[0008] As in the aforementioned rotating machinery abnormal characteristic detection system and method, the abnormal characteristic detection module defines the accessory as abnormal when the difference between the plurality of third quartiles is the largest.

[0009] As in the aforementioned rotating machinery abnormality characteristic detection system and method, the characteristic value distribution is a signal-to-noise ratio distribution, and the third quartile is calculated from the complex signal-to-noise ratio values ​​of the interval.

[0010] As described in the aforementioned rotating machinery abnormality characteristic detection system and method, the data processing module first calculates the standard deviation of the vibration data, and defines the portion of the vibration data greater than the standard deviation as an operating interval, and defines the portion of the vibration data less than the standard deviation as a standby interval, and defines the vibration data corresponding to the operating interval between two adjacent standby intervals as the vibration timing data.

[0011] As in the aforementioned rotating machinery abnormality characteristic detection system and method, the conversion module first converts each of the complex vibration time series data into complex intermediate data, and then converts the complex intermediate data into the complex conversion data.

[0012] As described in the aforementioned rotating machinery abnormality feature detection system and method, the conversion module uses a first algorithm to convert each of the complex vibration time series data into complex intermediate data, and the X-axis of the intermediate data is the frequency and the Y-axis is the signal-to-noise ratio value, wherein the signal-to-noise ratio value is the vibration amount of the vibration time series data divided by the standard deviation.

[0013] As in the aforementioned rotating machinery abnormality feature detection system and method, the first algorithm is Fourier transform, fast Fourier transform, wavelet transform or empirical mode decomposition.

[0014] As in the aforementioned rotating machinery abnormality feature detection system and method, the conversion module converts the complex intermediate data into the complex conversion data using a second algorithm, and the X-axis of the conversion data is the spacing and the Y-axis is the signal-to-noise ratio value.

[0015] As in the aforementioned rotating machinery abnormality feature detection system and method, the formula of the second algorithm is: , where d is the distance, v is the operating speed of the rotating machine, and f is the frequency.

[0016] As in the aforementioned rotating machinery abnormality characteristic detection system and method, the vibration data is mechanical motion vibration data or sound vibration data.

[0017] In summary, the disclosed rotating machinery abnormality feature detection system and method can be applied to the following scenarios and effectively locate abnormalities in accessories because the converted data uses pitch as the X-axis instead of frequency as the X-axis: (1) in the case of different operating speeds of the rotating machinery; (2) in the case of setting the pitch between the multiple teeth on the saw belt to be unevenly distributed in order to avoid resonance during operation. Simple diagram description

[0018] FIG1 is a system architecture diagram of the rotary machine abnormality detection system disclosed herein.

[0019] FIG. 2 is a flow chart of the disclosed method for detecting abnormal characteristics of a rotating machine.

[0020] FIG. 3 is a schematic diagram of vibration data of the present disclosure.

[0021] FIG. 4 is a schematic diagram of vibration timing data of the present disclosure.

[0022] FIG5 is a schematic diagram of the intermediate data of the present disclosure.

[0023] FIG6 is a schematic diagram of conversion data according to the present disclosure.

[0024] FIG. 7 is a schematic diagram of continuous complex conversion data according to the present disclosure.

[0025] FIG8 is a schematic diagram showing changes in the eigenvalue distribution of the present disclosure. Implementation Method

[0026] The following describes the implementation of the present disclosure through specific embodiments. People familiar with this technology can easily understand other advantages and functions of the present disclosure from the content disclosed in this specification, and can also implement or apply it through other different embodiments.

[0027] FIG1 is a system architecture diagram of a rotating machine abnormality feature detection system 1 disclosed herein. The rotating machine abnormality feature detection system 1 includes a sensing module 11 , a data processing module 12 , a conversion module 13 and an abnormality feature detection module 14 .

[0028] In this embodiment, the rotating machinery abnormality feature detection system 1 can be operated in a computer device having a processing unit and a storage unit. The processing unit can be a central processing unit (CPU), a microprocessor (MCU), a graphics processing unit (GPU), or an application specific integrated circuit (ASIC). The storage unit can be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, hard disk, soft disk, database, or a combination of the above components. The computer device can be a mobile phone, a tablet computer, a laptop computer, a desktop computer, a server, or a cloud server.

[0029] The data processing module 12, the conversion module 13, and the abnormal feature detection module 14 can be respectively program code segments, software, or firmware stored in a storage unit and can be executed by a processing unit, but the present disclosure is not limited to this. The data processing module 12, the conversion module 13, and the abnormal feature detection module 14 can also be implemented using other hardware or a combination of hardware and software.

[0030] Sensing module 11 is used to obtain vibration data from rotating machinery components. In this embodiment, sensing module 11 is specifically a sensor made of piezoelectric material (crystal or ceramic) or a conventional vibration sensor, but the present disclosure is not limited thereto. When sensing module 11 is a sensor made of piezoelectric material, the vibration data is acoustic vibration data. When sensing module 11 is a conventional vibration sensor, the vibration data is mechanical motion vibration data.

[0031] In one embodiment, the rotating machine may be, for example, an electric saw with a saw blade, in which case the corresponding accessory is a saw belt. In other embodiments, the rotating machine may be, for example, a power transmission device, a conveyor belt device, an electric drill device, or a vehicle device, in which case the corresponding accessories are gears, chains / belts, drill bits, or tires, although the present disclosure is not limited thereto.

[0032] Please refer to Figure 2. The disclosed method for detecting abnormal characteristics of rotating machinery can be executed by the aforementioned abnormal characteristics detection system 1 of rotating machinery. The method includes: in step S1, obtaining vibration data of a component of the rotating machinery; in step S2, processing the vibration data to generate complex vibration time series data; in step S3, converting each of the complex vibration time series data into complex intermediate data; in step S4, converting the complex intermediate data into complex converted data; in step S5, comparing the eigenvalue distributions in the complex converted data, and defining the component corresponding to the converted data with a change in the eigenvalue distribution as abnormal.

[0033] The following further describes the detailed operation of the rotating machinery abnormality detection system 1 in the order of the rotating machinery abnormality detection method. Technical details of the rotating machinery abnormality detection method that are identical to those of the rotating machinery abnormality detection system 1 will not be repeated.

[0034] FIG3 is a schematic diagram of vibration data 2 of a rotating machinery component obtained by sensing module 11. In one embodiment, sensing module 11 can link vibration data recorded for a fixed time period (e.g., 5 seconds) to form a continuous and complete set of vibration data 2, as shown in FIG3 . The X-axis of FIG3 represents time (minutes) and the Y-axis represents vibration intensity (G).

[0035] The data processing module 12 processes the vibration data 2 to generate multiple vibration time series data. In this embodiment, the data processing module 12 first calculates the standard deviation of the vibration data 2 and defines the portion of the vibration data 2 greater than the standard deviation as an operating interval 21, and defines the portion of the vibration data 2 less than the standard deviation as a standby interval 22. In this way, the operating interval 21 and the standby interval 22 can be used to distinguish the operation start and end points. For example, the standby interval 22 on the far left in FIG. 3 can be considered the operation start point, and the standby interval 22 second from the left in FIG. 3 can be considered the operation end point. The vibration data of the operating interval 21 between these two standby intervals 22 can be used as the vibration time series data. That is, the vibration data 2 shown in FIG3 can generate seven vibration time series data (for example, it can represent that an electric saw device with a sawing machine uses a saw belt to perform seven cutting operations). When the X-axis unit of each vibration time series data is expressed in minutes and the Y-axis unit is converted from the vibration amount (G) to the amplitude (G) - peak to peak, it can be expressed as vibration time series data 3 shown in FIG4.

[0036] In one embodiment, all vibration data 2 within the operating range 21 can be used as the vibration time series data 3, or the vibration data 2 within a certain period of time before and after the midpoint of the operating range 21 (for example, 5 minutes before and after) can be used as the vibration time series data 3, but the present disclosure is not limited to this.

[0037] The conversion module 13 converts each complex vibration time series data 3 into complex converted data 5. In this embodiment, the conversion module 13 first converts each complex vibration time series data 3 into complex intermediate data 4 (as shown in FIG. 5 ) using a first algorithm, and then converts the complex intermediate data 4 into complex converted data 5 (as shown in FIG. 6 ) using a second algorithm.

[0038] In this embodiment, the first algorithm is a Fourier transform, fast Fourier transform, wavelet transform, or empirical mode decomposition. Taking the fast Fourier transform as an example, the first algorithm can convert the vibration time series data 3 shown in Figure 4 from the time domain to the frequency domain, that is, from time (minutes) to frequency (Hz). Next, the vibration magnitude (G) on the Y axis of the vibration time series data 3 is divided by the standard deviation of the vibration data 2 to convert the vibration magnitude (G) on the Y axis of the vibration time series data 3 into a signal-to-noise ratio (dB). The intermediate data 4 is then displayed with the frequency (Hz) on the X axis and the signal-to-noise ratio (dB) on the Y axis, as shown in Figure 5.

[0039] In this embodiment, the formula of the second algorithm is , where d is the pitch (mm), v is the operating speed of the rotating machine (m / min), and f is the frequency (Hz). The converted data 5 is then displayed with the pitch (mm) on the X-axis and the signal-to-noise ratio (dB) on the Y-axis, as shown in Figure 6. The purpose of converting the vibration time series data 3 into converted data 5 with the pitch (mm) as the X-axis is to ensure a consistent comparison basis for each complex vibration time series data 3. If only fast Fourier transform is performed, only intermediate data 4 is obtained, as shown in Figure 5, with the X-axis representing the frequency. This may be impossible to compare when the X-axis represents inconsistent frequencies due to different operating speeds of the rotating machine. For example, when comparing the same peak in the vibration time series data 3, the peak may correspond to 60 Hz at slow operating speeds, while the peak may correspond to 120 Hz at fast operating speeds. This creates a lack of a consistent basis and makes comparison impossible. Therefore, after the vibration time series data 3 is converted into the conversion data 5 with the pitch (mm) as the X-axis by the second algorithm, when facing different operating speeds of the rotating machinery, the obtained complex conversion data 5 still has the same pitch and can be used as a benchmark for comparison.

[0040] The abnormal feature detection module 14 compares the eigenvalue distributions of the complex transformed data 51, 52, 53, 54, 55, 56, 57, and 58, and defines the component corresponding to the transformed data with a change in the eigenvalue distribution as abnormal. Specifically, the complex transformed data 51, 52, 53, 54, 55, 56, 57, and 58 can be presented as shown in FIG7 , so that the complex transformed data 51, 52, 53, 54, 55, 56, 57, and 58 have the same X-axis, that is, the same interval. The abnormal feature detection module 14 first compares the eigenvalue distributions of the transformed data 51 and the transformed data 52. Specifically, the eigenvalue distributions are signal-to-noise ratio distributions. The abnormal feature detection module 14 calculates the third quartile of the signal-to-noise ratio distributions corresponding to the same interval for the transformed data 51 and the transformed data 52, and then compares the third quartiles of the transformed data 51 and the transformed data 52 to see if there is a change. If the third quartiles of converted data 51 and converted data 52 have similar or identical signal-to-noise ratios, it is determined that no change has occurred (for example, this could indicate that the saw blade in a power saw machine is not damaged). If no change is found after comparing converted data 51 and converted data 52, the next round of comparison can be performed, comparing converted data 52 and converted data 53, and so on.

[0041] The following situations determine a change. For example, when comparing converted data 54 and converted data 55, as shown in Figure 8, at the interval of 22.1 to 22.2 mm, the third quartile 541 of converted data 54 and the third quartile 551 of converted data 55 have different signal-to-noise ratios. This indicates that converted data 55 has experienced change 6 (Figure 7), meaning that the distribution of the characteristic values ​​of converted data 55 has changed. In this case, the accessory corresponding to converted data 55 can be defined as abnormal (for example, a saw belt in a power saw with a sawing machine has damaged teeth). Subsequently, based on the distance corresponding to change 6 (e.g., 22.1 mm to 22.2 mm), the abnormal accessory location can be deduced. For example, if the accessory is a saw belt, the location of the damaged teeth can be deduced.

[0042] In one embodiment, an accessory is defined as abnormal when the difference between multiple third quartiles is the largest (i.e., the difference has the maximum value). In other embodiments, a range may be defined, and an accessory is defined as abnormal when the difference between multiple third quartiles falls within the range, or a threshold may be defined, and an accessory is defined as abnormal when the difference between multiple third quartiles exceeds the threshold. The present disclosure is not limited to this.

[0043] In other embodiments, in addition to using the third quartile, the first quartile or median of the transformed data may be calculated to determine whether the eigenvalue distribution has changed, but the present disclosure is not limited thereto.

[0044] The above embodiment compares a single conversion data with the next single conversion data, but the present disclosure is not limited to this. The characteristic value distribution of the integrated N-pieces of plural conversion data 51, 52, 53, 54, 55, 56, 57, and 58 can also be compared, where N is a natural number.

[0045] Specifically, the complex converted data 51, 52, 53, 54, 55, 56, 57, and 58 can be grouped into, for example, four groups to define complex converted data groups 71 and 75. In this case, the converted data groups 71 and 75 are adjacent to each other, and the converted data 51, 52, 53, and 54 included in the converted data group 71 do not overlap with the converted data 55, 56, 57, and 58 included in the converted data group 75. The eigenvalue distributions for the same intervals are calculated for each of the converted data groups 71 and 75, for example, the third quartile corresponding to the converted data 51, 52, 53, and 54, and the third quartile corresponding to the converted data 55, 56, 57, and 58. Because the converted data 55, 56, 57, and 58 all have a change point 6, the eigenvalue distribution of the converted data group 75 differs from that of the converted data group 71, and the difference reaches a maximum value. In this case, the accessory corresponding to the converted data 55 in the converted data group 75 can be defined as abnormal.

[0046] In one embodiment, the complex converted data 51, 52, 53, 54, 55, 56, 57, and 58 can be grouped into, for example, four complex converted data groups 71, 72, 73, 74, and 75. Conversion data group 71 includes conversion data 51, 52, 53, and 54; conversion data group 72 includes conversion data 52, 53, 54, and 55; conversion data group 73 includes conversion data 53, 54, 55, and 56; conversion data group 74 includes conversion data 54, 55, 56, and 57; and conversion data group 75 includes conversion data 55, 56, 57, and 58. The complex converted data 51, 52, 53, 54, 55, 56, 57, and 58 in the complex converted data groups 71, 72, 73, 74, and 75 partially overlap. Because conversion data 55 included in conversion data group 72 has a change at point 6, the eigenvalue distribution of conversion data group 72 has changed compared to that of conversion data group 71. To avoid misjudgments due to noise, the eigenvalue distributions of conversion data group 73 can be compared with those of conversion data group 71. Because conversion data 55 and 56 included in conversion data group 73 both have a change at point 6, the change in the eigenvalue distribution of conversion data group 73 compared to that of conversion data group 71 is greater than the change in the eigenvalue distribution of conversion data group 72 compared to that of conversion data group 71. In other words, the third quartile of conversion data group 73 is greater than the third quartile of conversion data group 72. Similarly, by comparing the eigenvalue distributions of conversion data group 74 with those of conversion data group 71, and then with those of conversion data group 75, we find that the third quartile of conversion data group 75 is greater than the third quartiles of conversion data groups 71, 72, 73, and 74, respectively. In this case, the accessory corresponding to conversion data 55 in conversion data group 75 can be defined as abnormal.

[0047] The above embodiment compares conversion data groups 72, 73, 74, and 75 with conversion data group 71, but the present disclosure is not limited to this. In other embodiments, conversion data groups 71, 72, 73, 74, and 75 can be compared in pairs, with the component corresponding to conversion data 55 in conversion data group 75 having the largest third quartile being defined as abnormal.

[0048] In summary, the disclosed rotating machinery abnormality feature detection system and method can be applied to the following scenarios and effectively locate abnormalities in accessories because the converted data uses pitch as the X-axis instead of frequency as the X-axis: (1) in the case of different operating speeds of the rotating machinery; (2) in the case of setting the pitch between the multiple teeth on the saw belt to be unevenly distributed in order to avoid resonance during operation.

[0049] The above embodiments are merely illustrative of the technical principles, features, and functions of this disclosure and are not intended to limit the scope of its application. Anyone skilled in the art may modify and alter the above embodiments without departing from the spirit and scope of this disclosure. However, any equivalent modifications and alterations resulting from the teachings of this disclosure are still covered by the following patent claims. The scope of protection of this disclosure is as set forth in the following patent claims.

[0050] 1: Rotating Mechanical Abnormal Feature Detection System 11: Sensing module 12: Data processing module 13:Conversion module 14: Abnormal feature detection module 2: Vibration data 21: Operation range 22: Standby zone 3: Vibration timing data 4: Intermediate data 5,51-58: Conversion data 541,551: third quartile 6: Changes 71-75: Convert data groups S1-S5: Steps

Claims

1. A rotating mechanical anomaly detection system, comprising: Sensing modules are used to acquire vibration data of components in rotating machinery; The data processing module is used to process the vibration data to generate complex vibration time series data; The conversion module converts each complex vibration time series data into complex converted data; and the anomaly detection module compares the feature value distribution in the complex converted data to define the component corresponding to the converted data whose feature value distribution changes as an anomaly; wherein, the anomaly detection module sequentially compares whether the feature value distribution of the same interval in any two adjacent complex converted data changes, or first defines the complex converted data into a group of N, and then sequentially compares whether the feature value distribution of the same interval in any two adjacent complex converted data groups changes, where N is a natural number.

2. The rotating mechanical anomaly detection system as described in claim 1, wherein, The change in the distribution of this characteristic value is due to the fact that the third and fourth quartiles of the complex numbers corresponding to the same interval in the complex transformed data or the complex transformed data group are different from each other.

3. The rotating mechanical anomaly detection system as described in claim 1, wherein, The complex transformation data in the groups may partially overlap or not overlap at all.

4. The rotating mechanical anomaly detection system as described in claim 2, wherein, The anomaly detection module defines the accessory as abnormal when the difference between the third and fourth quartiles of the complex number is the largest.

5. The rotating mechanical anomaly detection system as described in claim 4, wherein, The eigenvalue distribution is a signal-to-noise ratio distribution, and the third quartile is calculated from the complex signal-to-noise ratio values ​​of the interval.

6. The rotating mechanical anomaly detection system as described in claim 1, wherein, The data processing module first calculates the standard deviation of the vibration data, defines the portion of the vibration data that is greater than the standard deviation as the operating range, and defines the portion of the vibration data that is less than the standard deviation as the standby range. The vibration data of the operating range between two adjacent standby ranges is then defined as the vibration timing data.

7. The rotating mechanical anomaly detection system as described in claim 6, wherein, The conversion module first converts each complex vibration time series data into complex intermediate data, and then converts the complex intermediate data into the complex converted data.

8. The rotating mechanical anomaly detection system as described in claim 7, wherein, The conversion module uses a first algorithm to convert each complex vibration time series data into complex intermediate data, where the X-axis of the intermediate data is frequency and the Y-axis is signal-to-noise ratio, and the signal-to-noise ratio is the vibration amount of the vibration time series data divided by the standard deviation.

9. The rotating mechanical anomaly detection system as described in claim 8, wherein, The first algorithm is Fourier transform, fast Fourier transform, wavelet transform, or empirical mode decomposition.

10. The rotating mechanical anomaly detection system as described in claim 7, wherein, The conversion module uses a second algorithm to convert the complex intermediate data into the complex converted data, and the X-axis of the converted data is the spacing and the Y-axis is the signal-to-noise ratio value.

11. The rotating mechanical anomaly detection system as described in claim 10, wherein, The formula for the second algorithm is as follows: where d is the spacing, v is the operating speed of the rotating machine, and f is the frequency.

12. The rotating mechanical anomaly detection system as described in claim 1, wherein, The vibration data can be either mechanical motion vibration data or sound vibration data.

13. A method for detecting abnormal features of rotating machinery, comprising: The sensing module acquires vibration data of the components of rotating machinery; The data processing module processes the vibration data to generate complex vibration time series data; The conversion module converts each complex vibration time series data into complex converted data; and the anomaly detection module compares the feature value distribution in the complex converted data to define the component corresponding to the converted data whose feature value distribution changes as an anomaly; wherein, the anomaly detection module sequentially compares whether the feature value distribution of the same interval in any two adjacent complex converted data changes, or first defines the complex converted data into a group of N, and then sequentially compares whether the feature value distribution of the same interval in any two adjacent complex converted data groups changes, where N is a natural number.

14. The method for detecting abnormal features of rotating machinery as described in claim 13, wherein, The change in the distribution of this characteristic value is due to the fact that the third and fourth quartiles of the complex numbers corresponding to the same interval in the complex transformed data or the complex transformed data group are different from each other.

15. The method for detecting abnormal features of rotating machinery as described in claim 13, wherein, The complex transformation data in the groups may partially overlap or not overlap at all.

16. The method for detecting abnormal features of rotating machinery as described in claim 14, wherein, The anomaly detection module defines the accessory as abnormal when the difference between the third and fourth quartiles of the complex number is the largest.

17. The method for detecting abnormal features of rotating machinery as described in claim 16, wherein, The eigenvalue distribution is a signal-to-noise ratio distribution, and the third quartile is calculated from the complex signal-to-noise ratio values ​​of the interval.

18. The method for detecting abnormal features of rotating machinery as described in claim 13, wherein, The data processing module first calculates the standard deviation of the vibration data, defines the portion of the vibration data that is greater than the standard deviation as the operating range, and defines the portion of the vibration data that is less than the standard deviation as the standby range. The vibration data of the operating range between two adjacent standby ranges is then defined as the vibration timing data.

19. The method for detecting abnormal features of rotating machinery as described in claim 18, wherein, The conversion module first converts each complex vibration time series data into complex intermediate data, and then converts the complex intermediate data into the complex converted data.

20. The method for detecting abnormal features of rotating machinery as described in claim 19, wherein, The conversion module uses a first algorithm to convert each complex vibration time series data into complex intermediate data, where the X-axis of the intermediate data is frequency and the Y-axis is signal-to-noise ratio, and the signal-to-noise ratio is the vibration amount of the vibration time series data divided by the standard deviation.

21. The method for detecting abnormal features of rotating machinery as described in claim 20, wherein, The first algorithm is Fourier transform, fast Fourier transform, wavelet transform, or empirical mode decomposition.

22. The method for detecting abnormal features of rotating machinery as described in claim 19, wherein, The conversion module uses a second algorithm to convert the complex intermediate data into the complex converted data, and the X-axis of the converted data is the spacing and the Y-axis is the signal-to-noise ratio value.

23. The method for detecting abnormal features of rotating machinery as described in claim 22, wherein, The formula for the second algorithm is as follows: where d is the spacing, v is the operating speed of the rotating machine, and f is the frequency.

24. The method for detecting abnormal features of rotating machinery as described in claim 13, wherein, The vibration data can be either mechanical motion vibration data or sound vibration data.

Images