Ball Screw Fault Diagnosis Device and Fault Diagnosis Method

Abstract

Provided are an abnormality diagnosis device and an abnormality diagnosis method for a ball screw that allow for detection of abnormality such as damage or abrasion that has occurred in a screw shaft. This abnormality diagnosis device comprises: a vibration sensor (21) that detects vibration during operation of a ball screw; a filtering processing unit (23) that executes filtering processing for extracting a frequency band including at least a feature frequency corresponding to the natural vibration rate of the ball screw with respect to a vibration signal acquired using the vibration sensor (21); an envelope processing unit (241) that executes envelope processing of the vibration signal after the filtering processing; a frequency analysis processing unit (242) that executes fast Fourier transform processing on a time domain signal after the envelope processing; a peak hold and averaging processing unit (243) that executes peak hold and averaging processing of frequency domain data for each sampling period in the fast Fourier transform processing; and an abnormality determination unit (25) that determines abnormality of the ball screw on the basis of the frequency domain data after the peak hold and averaging processing.

Description

Technical Field

[0001] The present invention relates to an abnormal diagnosis apparatus and an abnormal diagnosis method for ball screws.

Background Art

[0002] Patent Document 1 discloses an abnormal diagnosis apparatus and an abnormal diagnosis method for mechanical equipment. In this Patent Document 1, the vibration signal of the mechanical equipment is filtered and envelope processing is performed, and an abnormal diagnosis is performed based on the magnitude (signal intensity) of the frequency spectrum of the characteristic frequency component corresponding to the natural vibration frequency of the bearing and the gear after fast Fourier transform (FFT) processing. In such vibration analysis processing using FFT, generally, the addition average processing of each data after FFT processing acquired in a plurality of periods corresponding to the required frequency resolution is performed to improve the SN ratio. On the other hand, Patent Document 2 discloses an abnormal determination method for a ball screw using a frequency component below the natural vibration frequency of a nut constituting the ball screw in the abnormal determination based on vibration during the operation of the ball screw.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] Damage and wear of the bearings and nuts of a ball screw can be determined by applying a vibration analysis process using FFT, as described in Patent Document 1, for example. However, determining damage and wear of the screw shaft of a ball screw requires moving the nut over the entire area of the screw shaft. Therefore, when applying a vibration analysis process using FFT, the data obtained multiple times after FFT processing is each data from a different position on the screw shaft. For this reason, the natural frequency (characteristic frequency component) generated due to an abnormality in the screw shaft is attenuated by the arithmetic averaging process after FFT processing, and there is a possibility that an abnormality such as damage or wear occurring in a part of the screw shaft cannot be detected.

[0005] The present invention has been made in consideration of the above-mentioned problems, and aims to provide an abnormality diagnosis device and an abnormality diagnosis method for a ball screw that can detect abnormalities such as damage and wear that have occurred in the screw shaft. [Means for solving the problem]

[0006] In order to achieve the above object, a ball screw abnormality diagnosis device according to one embodiment of the present invention comprises a vibration sensor that detects vibrations during operation of the ball screw, a filter processing unit that performs filtering on the vibration signal acquired by the vibration sensor to extract a frequency band that includes at least a characteristic frequency corresponding to the natural frequency of the ball screw, an envelope processing unit that performs envelope processing on the vibration signal after the filtering processing, a frequency analysis processing unit that performs fast Fourier transform processing on the time domain signal after the envelope processing, a peak hold averaging processing unit that performs peak hold averaging of frequency domain data for each sampling period in the fast Fourier transform processing, and an abnormality judgment unit that judges whether there is an abnormality in the ball screw based on the frequency domain data after the peak hold averaging processing.

[0007] With the above configuration, it is possible to execute a ball screw abnormality determination process using the maximum signal strength for each frequency line of the frequency domain data acquired for each sampling period while the ball screw is in operation, thereby detecting abnormalities such as damage or wear occurring in the screw shaft of the ball screw.

[0008] As a desirable aspect of the ball screw abnormality diagnosis device, it is preferable that the abnormality determination unit extracts a signal strength corresponding to a damage frequency of the screw shaft based on the rotational frequency of the screw shaft of the ball screw, and outputs an abnormality determination result for the screw shaft when the signal strength exceeds a predetermined threshold value.

[0009] This makes it possible to output a diagnosis result indicating that an abnormality has occurred in the screw shaft of the ball screw due to damage or wear.

[0010] In a preferred embodiment of the ball screw abnormality diagnosis device, the abnormality determination unit extracts the maximum value of the signal strength within a predetermined range that includes the damage frequency of the screw shaft of the ball screw, and if the maximum value exceeds a predetermined threshold value, outputs an abnormality determination result for the screw shaft.

[0011] This makes it possible to output a diagnosis result indicating that an abnormality has occurred in the screw shaft of the ball screw due to damage or wear.

[0012] A method for diagnosing an abnormality in a ball screw according to one embodiment of the present invention includes a first step of detecting vibrations during operation of the ball screw, a second step of performing a filtering process on the vibration signal acquired in the first step to extract a frequency band including at least a characteristic frequency corresponding to the natural frequency of the ball screw, a third step of performing envelope processing on the vibration signal after the filtering process, a fourth step of performing fast Fourier transform processing on the time domain signal after the envelope processing, a fifth step of performing peak hold averaging processing of frequency domain data for each sampling period in the fast Fourier transform processing, and a sixth step of determining an abnormality in the ball screw based on the frequency domain data after the peak hold averaging processing.

[0013] With the above configuration, it is possible to execute the abnormality determination process of the ball screw by using the maximum signal intensity for each frequency line of the frequency domain data acquired for each sampling period during the operation of the ball screw. Thereby, it is possible to detect abnormalities such as damage and wear occurring on the screw shaft.

[0014] As a desirable aspect of the ball screw abnormality diagnosis method, in the sixth step, based on the rotational frequency of the screw shaft of the ball screw, the signal intensity corresponding to the damage frequency of the screw shaft is extracted, and when the signal intensity exceeds a predetermined threshold value, it is preferable to output the abnormality determination result of the screw shaft.

[0015] Thereby, it is possible to output a diagnosis result that determines that an abnormality due to damage or wear has occurred on the screw shaft of the ball screw.

[0016] As a desirable aspect of the ball screw abnormality diagnosis method, in the sixth step, the maximum value of the signal intensity within a predetermined range including the damage frequency of the screw shaft of the ball screw is extracted, and when the maximum value exceeds a predetermined threshold value, it is preferable to output the abnormality determination result of the screw shaft.

[0017] Thereby, it is possible to output a diagnosis result that determines that an abnormality due to damage or wear has occurred on the screw shaft of the ball screw.

Advantages of the Invention

[0018] According to the present invention, an abnormality diagnosis apparatus and an abnormality diagnosis method for a ball screw capable of detecting abnormalities such as damage and wear occurring on the screw shaft can be obtained.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5A

Figure 5B

Figure 6

Figure 7

Figure 8

Figure 9

DETAILED DESCRIPTION OF THE INVENTION

[0020] Hereinafter, embodiments for carrying out the invention (hereinafter referred to as embodiments) will be described in detail with reference to the drawings. It should be noted that the present invention is not limited by the following embodiments. Further, the constituent elements in the following embodiments include those that can be easily assumed by those skilled in the art, substantially the same ones, and those within the so - called equivalent range. Furthermore, the constituent elements disclosed in the following embodiments can be combined as appropriate.

[0021] FIG. 1 is a diagram showing an example of the schematic configuration of an abnormality diagnosis system for a ball screw. The ball screw 1 has a screw shaft 11 and a nut 12 that is slidably fitted over the screw shaft 11 via a plurality of rolling elements (not shown).

[0022] Both ends of the screw shaft 11 are rotatably supported by bearings 13 and 14 respectively, and one end side (the right side in FIG. 1) of the screw shaft 11 is connected to the output shaft of the driving motor 15. As the screw shaft 11 rotates, the nut 12 linearly moves in the axial direction of the screw shaft 11 (the direction of the arrow shown in FIG. 1) between the A end and the B end shown in FIG. 1.

[0023] In the present disclosure, the abnormality diagnosis device 2 for the ball screw 1 according to the embodiment executes vibration analysis processing during the operation of the ball screw 1 to diagnose damage and wear of the screw shaft 11. Specifically, when executing the abnormality diagnosis process for the ball screw 1 according to the embodiment, the abnormality diagnosis device 2 outputs a drive control command and a rotation control command for the motor 15 to the drive control device 3. Based on the control command from the abnormality diagnosis device 2, the drive control device 3 drives the motor 15 to operate the ball screw 1 and moves the nut 12 over the entire area of the screw shaft 11. The drive control device 3 outputs the rotation frequency of the motor 15, that is, the rotation frequency of the screw shaft 11 of the ball screw 1, to the abnormality diagnosis device 2.

[0024] FIG. 2 is a block diagram showing an example of the abnormality diagnosis device for the ball screw according to the embodiment. As shown in FIG. 2, the abnormality diagnosis device 2 for the ball screw according to the embodiment includes a vibration sensor 21, an AD conversion processing unit 22, a filter processing unit 23, a vibration analysis processing unit 24, and an abnormality determination unit 25.

[0025] The vibration sensor 21 is installed, for example, on the nut 12 of the ball screw 1 shown in FIG. 1. In the present disclosure, the vibration sensor 21 is exemplified by an acceleration sensor such as an acceleration pickup. The acceleration detection direction by the vibration sensor 21 may be the axial direction of the screw shaft 11 (the direction of the arrow shown in FIG. 1), or may be a direction orthogonal to the axial direction of the screw shaft 11. Further, the installation location of the vibration sensor 21 may be any position where the vibration during the operation of the ball screw 1 can be detected, and is not limited to the nut 12. Furthermore, the vibration sensor 21 is not limited to an acceleration sensor. The vibration sensor 21 may be, for example, a displacement sensor or a sound pressure sensor such as a microphone.

[0026] The AD conversion processing unit 22 converts the vibration signal detected by the vibration sensor 21 into digital data.

[0027] The filter processing unit 23 performs a predetermined filtering process on the vibration signal converted into digital data. Specifically, the filter processing unit 23 extracts a frequency band including characteristic frequencies corresponding to various natural frequencies of the ball screw 1. As the filter processing unit 23, for example, a low-pass filter is exemplified, but it is not limited thereto. The filter processing unit 23 may be a high-pass filter, a band-pass filter, or a band-stop filter as long as it can extract a frequency band including characteristic frequencies corresponding to various natural frequencies of the ball screw 1.

[0028] In the present disclosure, the frequency of the natural vibration generated due to damage or wear of the screw shaft 11 of the ball screw 1 is defined as the characteristic frequency. Hereinafter, the frequency of the natural vibration generated due to damage or wear of the screw shaft 11 of the ball screw 1 is also referred to as the "damage frequency of the screw shaft 11".

[0029] The damage frequency of the screw shaft 11 can be expressed as a value obtained by multiplying the relative rotational frequency fi of the screw shaft 11 with respect to the revolution frequency of the rolling elements of the nut 12 by the number of rolling pairs Z per lead (= Z × fi, hereinafter also simply referred to as "Zfi"). The damage frequency Zfi of the screw shaft 11 is given by the following formula (1) when the axial rotational frequency of the screw shaft 11 is fr, the rolling element diameter is Dw, the center circle diameter of the rolling elements is Da, the screw groove contact angle is α, and the lead angle is β. The axial rotational frequency fr of the screw shaft 11 can be calculated from the rotational frequency of the motor 15 (the rotational frequency of the screw shaft 11 of the ball screw 1).

[0030]

Equation

[0031] The rotational frequency of the motor 15 (the rotational frequency of the screw shaft 11 of the ball screw 1) corresponds to the moving speed of the nut 12 when performing the abnormality diagnosis process of the present disclosure.

[0032] In the present disclosure, the abnormality determination unit 25 preliminarily holds a signal intensity threshold corresponding to the damage frequency Zfi of the screw shaft 11 represented by the above formula (1). The abnormality determination unit 25 executes a threshold determination process between the signal intensity threshold corresponding to the damage frequency Zfi of the screw shaft 11 held in advance and the signal intensity corresponding to the damage frequency Zfi acquired by the vibration analysis processing unit 24. Hereinafter, the configuration and operation of the vibration analysis processing unit 24 according to the embodiment will be described.

[0033] The vibration analysis processing unit 24 according to the embodiment includes an envelope processing unit 241, a frequency analysis processing unit 242, and a peak hold average processing unit 243. The envelope processing unit 241 performs envelope processing (envelope detection processing) on the vibration signal filtered by the filter processing unit 23. Specifically, the envelope processing unit 241 executes envelope processing of the vibration signal using, for example, Hilbert transform. The envelope processing unit 241 may be configured to execute envelope processing of the vibration signal by absolute value detection, for example.

[0034] The frequency analysis processing unit 242 performs frequency spectrum analysis processing of the time domain signal. Specifically, the frequency analysis processing unit 242 executes fast Fourier transform (FFT) processing on the time domain signal after envelope processing. FIG. 3 is an image diagram schematically showing a vibration signal acquired in the abnormality diagnosis process according to the embodiment. In FIG. 3, the horizontal axis represents time, and the vertical axis represents the magnitude (acceleration a) of the vibration signal.

[0035] In FIG. 3, the period for data acquisition in the abnormality diagnosis process according to the embodiment is set as T. The frequency resolution in the FFT process is determined by the sampling rate (sampling frequency) and the block size (number of sampling data). The sampling period (time window length) D of the FFT process in the data acquisition period T is defined as the reciprocal of the frequency resolution in the FFT process.

[0036] In the abnormal diagnosis process of bearings, gears, etc., it is common to perform an addition average process on the frequency domain data (signals after FFT processing) acquired multiple times during the data acquisition period T to improve the SNR. However, in the determination of damage or wear of the screw shaft 11 of the ball screw 1 to be abnormally diagnosed in the present disclosure, since it is necessary to move the nut 12 over the entire area of the screw shaft 11, the frequency domain data acquired multiple times during the data acquisition period T are respectively vibration data acquired at different positions of the screw shaft 11. For this reason, the signal intensity corresponding to the damage frequency Zfi of the screw shaft 11 may be attenuated by the addition average process, and there is a possibility that abnormalities such as damage and wear occurring in a part of the screw shaft 11 cannot be detected. Hereinafter, the concept of the vibration analysis process according to the present disclosure will be described.

[0037] FIG. 4 is a diagram showing a specific example of a vibration signal acquired in the abnormal diagnosis process according to the embodiment. In the example shown in FIG. 4, an example in which the nut 12 is reciprocated three times between the A end and the B end shown in FIG. 1 during the data acquisition period T is shown.

[0038] In the example shown in FIG. 4, during the periods T21 and T31, the magnitude (acceleration a) of the vibration signal becomes an abnormal value at the C point close to the A end. There is a possibility that damage or wear has occurred in the screw shaft 11 at this C point.

[0039] FIGS. 5A and 5B are schematic diagrams showing an example of frequency domain data acquired by FFT processing. FIG. 6 is a schematic diagram showing an example of frequency domain data after addition average processing. FIG. 7 is a schematic diagram showing an example of frequency domain data after peak hold average processing. In FIGS. 5A, 5B, 6, and 7, the horizontal axis represents frequency, and the vertical axis represents the signal intensity of the frequency domain data.

[0040] In Fig. 5A, frequency domain data obtained when the nut 12 passes through the damaged portion of the screw shaft 11 is illustrated. When the vibration signals passing through the intervals T11A, T21A, and T31A in Fig. 4 are subjected to FFT processing, frequency domain data as shown in Fig. 5A is obtained. As shown in Fig. 5A, in the frequency domain data obtained when the nut 12 passes through the damaged portion of the screw shaft 11, the signal intensities of the damage frequency Zfi of the screw shaft 11, the second harmonic component 2Zfi, the third harmonic component 3Zfi, the fourth harmonic component 4Zfi, etc. of the damage frequency Zfi become high. However, in the frequency domain data obtained when the nut 12 passes through a portion where there is no damage or where the wear is minor, as shown in Fig. 5B, the signal intensities of the frequency components corresponding to the damage frequency Zfi of the screw shaft 11 and its harmonic components 2Zfi, 3Zfi, 4Zfi become low. Note that Fig. 5B illustrates a plurality of frequency domain data obtained when the nut 12 passes through a plurality of portions where there is no damage or where the wear is minor. When the vibration signals passing through the intervals T11B, T21B, and T31B in Fig. 4 are subjected to FFT processing, frequency domain data as shown in Fig. 5B is obtained.

[0041] In such a configuration for performing abnormality diagnosis such as damage and wear of the screw shaft 11 of the ball screw 1, as shown in Fig. 6, when the frequency domain data acquired multiple times is subjected to addition and averaging processing, the signal intensities of the damage frequency Zfi of the screw shaft 11 and its harmonic components 2Zfi, 3Zfi, 4Zfi, etc. may attenuate, and there is a possibility that abnormalities such as damage and wear occurring in a part of the screw shaft 11 cannot be detected.

[0042] When only a part of the screw shaft of the ball screw 1 is damaged, if frequency domain data is acquired multiple times, most of the frequency domain data is obtained when the nut 12 passes through the undamaged portion. That is, most of the frequency domain data has a low signal intensity as shown in Fig. 5B. When addition and averaging processing is performed, the signal intensity after the averaging process may attenuate due to the influence of the frequency domain data with a low signal intensity, and there is a possibility that abnormalities such as damage and wear occurring in a part of the screw shaft 11 cannot be detected.

[0043] For this reason, in the present disclosure, a peak hold averaging processor 243 that performs peak hold averaging of frequency domain data after FFT processing is provided downstream of a frequency analysis processor 242 that performs frequency spectrum analysis of a time domain signal.

[0044] Specifically, the peak hold average processing unit 243 acquires the maximum value for each frequency (frequency line) for which spectral data, determined by the frequency resolution of the frequency domain data after FFT processing, should be generated, and generates the spectral data shown in Fig. 7. This makes it possible to detect the signal strength of the damage frequency Zfi and each harmonic component 2Zfi, 3Zfi, 4Zfi, etc., that occurs due to damage or wear that has occurred in a part of the screw shaft 11.

[0045] The abnormality diagnosis process for the ball screw 1 according to the embodiment will be described below. Fig. 8 is a flowchart showing an example of the abnormality diagnosis process for the ball screw according to the embodiment.

[0046] In the abnormality diagnosis system for the ball screw 1 shown in FIG. 1 , the abnormality diagnosis device 2 according to the embodiment executes a vibration analysis process during operation of the ball screw 1, as described above, to diagnose damage and wear on the screw shaft 11. Specifically, when executing the abnormality diagnosis process for the ball screw 1 according to the embodiment, the abnormality diagnosis device 2 outputs a drive control command and a rotation control command for the motor 15 to the drive control device 3. Based on the control command from the abnormality diagnosis device 2, the drive control device 3 drives the motor 15 to operate the ball screw 1 and moves the nut 12 over the entire range of the screw shaft 11 at a speed corresponding to the damage frequency Zfi of the screw shaft 11 expressed by the above equation (1). In other words, the drive control device 3 rotates the motor 15 (screw shaft 11) so that the shaft rotation frequency fr corresponds to the damage frequency Zfi of the screw shaft 11 expressed by the above equation (1) over the entire range of the screw shaft 11. Specifically, the drive control device 3 reciprocates the nut 12 multiple times between ends A and B shown in FIG. 1, as shown in FIG. 4 , for example.

[0047] The vibration sensor 21 detects the vibration during the operation of the ball screw 1. The vibration signal detected by the vibration sensor 21 is converted into digital data by the AD conversion processing unit 22 and acquired (step S100).

[0048] The filter processing unit 23 executes a predetermined filtering process on the vibration signal acquired by the vibration sensor 21 (step S200). Specifically, the filter processing unit 23 extracts a frequency band including at least characteristic frequencies corresponding to various natural frequencies of the ball screw 1.

[0049] The envelope processing unit 241 executes an envelope process on the vibration signal after the filtering process (step S300). The frequency analysis processing unit 242 executes a fast Fourier transform (FFT) process on the time-domain signal after the envelope process and performs frequency spectrum analysis processing (step S400).

[0050] For the peak hold average processing unit 243, during the data acquisition period T (the operating period of the ball screw 1 in the abnormality diagnosis process according to the embodiment), the frequency domain data after the FFT process is sequentially input for each sampling period (time window length) D defined by the reciprocal of the frequency resolution in the FFT process. The peak hold average processing unit 243 executes peak hold average processing on the frequency domain data acquired in each sampling period D (step S500). FIG. 9 is a sub flowchart showing an example of the peak hold average processing.

[0051] In FIG. 9, n indicates an integer up to the total number N of the frequency domain data acquired during the data acquisition period T. Also, in FIG. 9, m indicates an integer up to the total number M of the frequency lines determined by the frequency resolution of the frequency domain data after the FFT process. Also, Sigfm indicates the signal intensity for each frequency line m.

[0052] The peak hold averaging processor 243 resets the number n of the frequency domain data acquired during the data acquisition period T (n=0, step S501), and determines whether n=N-1 (step S502). If n=N-1 is not true (step S502; No), the peak hold averaging processor 243 increments the number n of the frequency domain data (n=n+1, step S503).

[0053] Next, the peak hold average processing unit 243 resets the frequency line number m and the maximum value Sigfmmax of the signal strength Sigfm on the frequency line m (m=0, Sigfmmax=0, step S504), and determines whether m=M-1 (step S505). If m=M-1 is not true (step S505; No), the frequency line number m is incremented (m=m+1, step S506).

[0054] When the frequency domain data of the sampling period Dn corresponding to the number n is input, the peak hold averaging processor 243 determines whether the signal strength Sigfm on the frequency line m exceeds the maximum value Sigfmmax (step S507).

[0055] If the signal strength Sigfm exceeds the maximum value Sigfmmax (Sigfm>Sigfmmax, step S507; Yes), the signal strength Sigfm is set as the maximum value Sigfmmax on frequency line m (step S508), and the process returns to step S505, and the processes from step S505 onwards are repeatedly executed.

[0056] If the signal strength Sigfm is equal to or less than the maximum value Sigfmmax (Sigfm≦Sigfmmax, step S507; No), the process returns to step S505, and the processes from step S505 onwards are repeatedly executed.

[0057] In the process of step S505, when m = M - 1 (step S505; Yes), the process returns to step S502, and the processes after step S502 are repeatedly executed. In the process of step S502, when n = N - 1 (step S502; Yes), the process returns to the abnormality diagnosis process shown in FIG. 8. By the peak hold average process shown in FIG. 9 described above, the maximum value for each frequency line determined by the frequency resolution of the frequency domain data after the FFT process can be obtained, and the spectrum data shown in FIG. 7 can be generated.

[0058] Returning to FIG. 8, the abnormality determination unit 25 calculates the axial rotation frequency fr of the screw shaft 11 of the ball screw 1 based on the rotation frequency of the motor 15 (screw shaft 11), and extracts the signal intensity SigZfi corresponding to the damage frequency Zfi of the screw shaft 11 represented by the above formula (1) (step S600), and executes a comparison determination process with a signal intensity threshold value Sigth held in advance (step S700). Specifically, the abnormality determination unit 25 determines whether or not the signal intensity SigZfi extracted in the process of step S600 is less than or equal to the signal intensity threshold value Sigth.

[0059] In the present disclosure, the signal intensity SigZfi corresponding to the damage frequency Zfi of the screw shaft 11 may be the signal intensity that becomes the maximum value within a predetermined range including the damage frequency Zfi of the screw shaft 11. Specifically, the signal intensity SigZfi corresponding to the damage frequency Zfi of the screw shaft 11 is preferably, for example, the signal intensity that becomes the maximum value within the range of not less than 0.9Zfi and not more than Zfi.

[0060] When the signal intensity SigZfi extracted in the process of step S600 is less than or equal to the signal intensity threshold value Sigth (SigZfi ≤ Sigth, step S700; Yes), it is determined that there is no abnormality due to damage or wear of the screw shaft 11 of the ball screw 1 to be the abnormality diagnosis target in the present disclosure (step S800A), the diagnosed result of the normal determination is output (step S900), and the abnormality diagnosis process is terminated.

[0061] If the signal intensity SigZfi extracted in the process of step S600 exceeds the signal intensity threshold Sigth (SigZfi > Sigth, step S700; No), it is abnormally determined that an abnormality such as damage or wear has occurred in the screw shaft 11 of the ball screw 1 to be diagnosed for abnormality in the present disclosure (step S800B), the diagnosed result of the abnormality determination is output (step S900), and the abnormality diagnosis process is terminated.

[0062] As described above, according to the abnormality diagnosis device 2 and the abnormality diagnosis method of the ball screw 1 according to the embodiment of the present disclosure, it is possible to detect an abnormality such as damage or wear occurring in the screw shaft 11 of the ball screw 1.

Explanation of reference numerals

[0063] 1 Ball screw 2 Abnormality diagnosis device 3 Drive control device 11 Screw shaft 12 Nut 13, 14 Bearings 15 Motor 21 Vibration sensor 22 AD conversion processing unit 23 Filter processing unit 24 Vibration analysis processing unit 25 Abnormality determination unit 241 Envelope processing unit 242 Frequency analysis processing unit 243 Peak hold average processing unit

Claims

1. A vibration sensor that detects vibration during the operation of a ball screw, a filter processing unit that executes a filtering process for extracting a frequency band including at least a characteristic frequency corresponding to the natural frequency of the ball screw from the vibration signal acquired by the vibration sensor, an envelope processing unit that executes an envelope process on the vibration signal after the filtering process, a frequency analysis processing unit that executes a fast Fourier transform process on the time domain signal after the envelope process, a peak hold average processing unit that sequentially receives the frequency domain data after the frequency analysis process for each sampling period defined by the reciprocal of the frequency resolution in the fast Fourier transform process, and executes peak hold average processing on the frequency domain data acquired in each sampling period, an abnormality determination unit that determines an abnormality of the ball screw based on the frequency domain data after the peak hold average processing, and comprising an abnormality diagnosis device for a ball screw.

2. The peak hold average processing unit acquires the maximum value of the signal intensity acquired in each sampling period for each frequency line determined by the frequency resolution of the frequency domain data after the fast Fourier transform process, and generates frequency domain data based on the maximum value of the signal intensity for each frequency line, The abnormality diagnosis device for a ball screw according to Claim 1.

3. The abnormality determination unit extracts the signal intensity corresponding to the damage frequency of the screw shaft based on the rotational frequency of the screw shaft of the ball screw, and outputs an abnormality determination result of the screw shaft when the signal intensity exceeds a predetermined threshold value, The abnormality diagnosis device for a ball screw according to Claim 2.

4. The abnormality determination unit extracts the maximum value of the signal intensity within a predetermined range including the damage frequency of the screw shaft of the ball screw, and outputs an abnormality determination result of the screw shaft when the maximum value exceeds a predetermined threshold value, The abnormality diagnosis device for a ball screw according to Claim 2.

5. A first step of detecting vibration during the operation of a ball screw, a second step of executing a filtering process for extracting a frequency band including at least a characteristic frequency corresponding to the natural frequency of the ball screw from the vibration signal acquired in the first step, a third step of executing an envelope process on the vibration signal after the filtering process, a fourth step of executing a fast Fourier transform process on the time domain signal after the envelope process, For each sampling period defined by the reciprocal of the frequency resolution in the fast Fourier transform process, frequency-domain data after frequency analysis processing is sequentially input, and a fifth step of performing peak hold average processing on the frequency-domain data obtained in each sampling period; A sixth step of performing abnormality determination of the ball screw based on the frequency-domain data after the peak hold average processing; having An abnormality diagnosis method for a ball screw.

6. In the fifth step, For each frequency line determined by the frequency resolution of the frequency-domain data after the fast Fourier transform process, the maximum value of the signal intensity obtained in each sampling period is acquired, and frequency-domain data is generated based on the maximum value of the signal intensity for each frequency line. The abnormality diagnosis method for a ball screw according to claim 5.

7. In the sixth step, Based on the rotational frequency of the screw shaft of the ball screw, the signal intensity corresponding to the damage frequency of the screw shaft is extracted, and when the signal intensity exceeds a predetermined threshold value, the abnormality determination result of the screw shaft is output. The abnormality diagnosis method for a ball screw according to claim 6.

8. In the sixth step, The maximum value of the signal intensity within a predetermined range including the damage frequency of the screw shaft of the ball screw is extracted, and when the maximum value exceeds a predetermined threshold value, the abnormality determination result of the screw shaft is output. The abnormality diagnosis method for a ball screw according to claim 6.

Images