Rope wheel wear early warning method based on vibration signal

Abstract

The application discloses a kind of based on vibration signal's rope wheel wear early warning method, vibration data of rope wheel is collected according to preset period, according to the vibration data of each cycle collected respectively calculates its vibration energy;When total cycle n is not less than d+1, according to the vibration energy judges the wear stage when the n-dth cycle of rope wheel, wherein d≥3;The wear stage is divided into running-in wear stage, stable wear stage and severe wear stage;In the case where wear early warning threshold is not set, when the wear stage of the n-dth cycle of rope wheel is stable wear stage or severe wear stage, set wear early warning threshold;Judge whether the vibration energy of current cycle exceeds the wear early warning threshold, if exceed, send wear early warning signal.The sensor of the method of the application is easy to install, and automatic monitoring is easily realized.

Description

Technical Field

[0001] This invention relates to the field of rotating component fault detection technology, and specifically to a method for early warning of rope wheel wear based on vibration signals. Background Technology

[0002] The traction sheave and wire rope play a crucial role in the safe operation of an elevator. During operation, the traction sheave (hereinafter referred to as the rope sheave) and the wire rope come into contact with each other, generating friction that pulls the wire rope. Over time, friction causes wear on the rope sheave, reducing the diameter of the wire rope and lowering the coefficient of friction. This can lead to potential dangers such as insufficient traction force and wire rope breakage, affecting the safety of the elevator.

[0003] Currently, methods for detecting elevator sheave wear mainly include electromagnetic methods, machine vision methods, and current methods. For example, Chinese patent document CN201911411098.9 discloses an elevator sheave condition early warning method based on magnetic detection and current methods. This early warning method has two main shortcomings: firstly, the sensor is inconvenient to install; for example, a strong magnetic sensor may cause the wire rope to come into contact with the device, leading to unnecessary wear on the wire rope. Secondly, this method requires manual acquisition of parameters such as the number and surface area of ​​the elevator wire rope, which is not automated enough. Summary of the Invention

[0004] The technical problem to be solved by this invention is how to provide a rope and wheel wear early warning technology solution that is easy to install and can achieve automated monitoring.

[0005] To address the aforementioned technical problems, this invention provides a method for early warning of rope wheel wear based on vibration signals. Vibration data of the rope wheel is collected at preset intervals, and the vibration energy is calculated based on the vibration data collected in each interval. When the total collection period n is not less than d+1, the wear stage of the rope wheel in the nd-th interval is determined based on the vibration energy, where d≥3. The wear stages are divided into a break-in wear stage, a stable wear stage, and a severe wear stage.

[0006] If no wear warning threshold is set, when the wear stage of the pulley is a stable wear stage or a severe wear stage in the nd cycle, a wear warning threshold is set; it is determined whether the vibration energy of the current cycle exceeds the wear warning threshold, and if it does, a wear warning signal is issued.

[0007] Preferably, the method for determining the wear stage of the sheave in the nd-th cycle based on the vibration energy is as follows: Plot the vibration energy data from the nd-th cycle to the current cycle on a two-dimensional coordinate system, with the vertical axis representing the vibration energy and the horizontal axis representing the time period for collecting the sheave vibration data; use the least squares method to fit a univariate function curve to the (d+1)-th set of vibration energy data to obtain the slope k of the function; if k < -k thr If -k thr <k<k thr If k > k, then the wear stage is determined to be the stable wear stage; thr If k is the wear stage, then the wear stage is determined to be the severe wear stage; where k thr Given a threshold, k thr >0.

[0008] Preferably, the method for setting the wear warning threshold is as follows: when the wear stage of the sheave in the nd cycle is the stable wear stage, E thr =β1E b E thr β1 is the first warning coefficient, and E is the wear warning threshold. b The baseline value is E; when the sheave is in the severe wear stage during the nd cycle, E thr =β2E b E thr β2 is the wear warning threshold, and E is the second warning coefficient. b This is the baseline value.

[0009] Preferably, the first warning coefficient β1 is greater than the second warning coefficient β2.

[0010] Preferably, when the wear stage of the sheave is the stable wear stage in the nd cycle, the baseline value E is... b It is the average vibrational energy from the nd-th cycle to the current cycle.

[0011] Preferably, when the wear stage of the sheave is the severe wear stage in the nd cycle, the baseline value E b Let be the vibrational energy of the nd period.

[0012] Preferably, the vibration data of the rope pulley is collected every 30 days as the preset cycle.

[0013] Preferably, the sheave is at least one of a traction sheave, guide sheave, anti-rope sheave, or compensating sheave in the elevator system.

[0014] Preferably, the vibration data of the sheave is collected during the section of the elevator running at a constant speed from the end floor to the end floor.

[0015] Preferably, the method for calculating vibration energy based on the vibration data is as follows: determine the frequency band for calculating vibration energy, sum the time-domain signals of the vibration data in a single direction within the frequency band after processing, and calculate the vibration energy by performing an arithmetic square root.

[0016] Preferably, the single direction is a radial direction relative to the plane of rotation of the sheave.

[0017] Preferably, the method for determining the frequency band for calculating vibration energy is as follows: the time domain signal of each axis in each frequency band of vibration data is obtained by harmonic wavelet packet transform; the sum of the absolute values ​​of the correlation coefficients between each axis is calculated based on the time domain signal of each axis in each frequency band; and the frequency band with the largest sum of the absolute values ​​of the correlation coefficients is determined as the frequency band for calculating vibration energy.

[0018] Compared with existing technologies, the present invention allows for easy sensor installation without affecting the normal operation of the elevator, and also eliminates the need to obtain elevator structural parameters, thus facilitating automated monitoring. Attached Figure Description

[0019] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0020] Figure 1 This is a schematic diagram of harmonic wavelet packet transform decomposition of vibration data in Example 1;

[0021] Figure 2 This is a schematic diagram of the correlation coefficients between axes in each frequency band after harmonic wavelet packet transformation of the vibration data in Example 1;

[0022] Figure 3 This is a schematic diagram of the wear stages and wear warning of the rope pulley in Example 1;

[0023] Figure 4 This is a schematic diagram of the wear stages and wear warning of the rope pulley in Example 2;

[0024] Figure 5 This is a schematic diagram of the wear stages and wear warning of the rope pulley in Example 3;

[0025] Figure 6 This is a schematic diagram of the correlation coefficients between axes in each frequency band after harmonic wavelet packet transformation of vibration data in Example 4. Detailed Implementation

[0026] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can fully understand other advantages and technical effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through different specific embodiments, and the details in this specification can also be applied based on different viewpoints, with various modifications or changes made without departing from the overall design concept of the invention. It should be noted that, unless otherwise specified, the following embodiments and features can be combined with each other. The following exemplary embodiments of the present invention can be implemented in many different forms and should not be construed as being limited to the specific embodiments set forth herein. It should be understood that these embodiments are provided to make the disclosure of the present invention thorough and complete, and to fully convey the technical solutions of these exemplary embodiments to those skilled in the art.

[0027] Example 1

[0028] In this embodiment, a sheave in an elevator system is used as an example. The sheave in an elevator system can be, for example, a traction sheave, guide sheave, anti-corrosion sheave, or compensating sheave. A vibration sensor can be installed at the base of the sheave accessory that needs monitoring to collect vibration data. The vibration sensor can be a triaxial vibration sensor, a biaxial or uniaxial vibration sensor, but a triaxial vibration sensor is preferred.

[0029] This embodiment provides a method for early warning of rope wheel wear based on vibration signals. Vibration data of the rope wheel is collected at preset cycles, and the vibration energy is calculated based on the vibration data collected in each cycle. When the total collection cycle n is not less than d+1, the wear stage of the rope wheel in the nd cycle is determined based on the vibration energy, where d≥3. The wear stages are divided into running-in wear stage, stable wear stage and severe wear stage.

[0030] If no wear warning threshold is set, when the wear stage of the pulley is a stable wear stage or a severe wear stage in the nd cycle, a wear warning threshold is set; it is determined whether the vibration energy of the current cycle exceeds the wear warning threshold, and if it does, a wear warning signal is issued.

[0031] The preset period can be regular or irregular, with regular data collection being preferred, such as daily, weekly, monthly, or quarterly. This embodiment uses the vibration data of the rope sheave collected every 30 days as the preset period, and a sampling frequency f is installed on the traction sheave base. s It is a 5120Hz triaxial vibration sensor.

[0032] During each data collection, the vibration data of the sheave is collected for the section of the elevator running at a constant speed from the end floor to the end floor.

[0033] To calculate vibration energy based on the collected vibration data, the frequency band for calculating the vibration energy must first be determined. The method for determining the frequency band is as follows:

[0034] For a triaxial vibration sensor, harmonic wavelet packet transform is performed on the vibration data x(i), y(i), and z(i) in the x, y, and z axes. Taking the x-axis vibration data x(i) as an example, as follows... Figure 1 As shown, harmonic wavelet packet transform decomposition is performed. The highest analysis frequency of the signal is... The signal is decomposed into j (j = 0, 1, 2, 3, ...) frequency bands, and the bandwidth B of each band can be expressed as B = 2. -j f h s is the number of frequency bands, s = 2 j The upper limit m and lower limit n of each analysis frequency band are respectively

[0035]

[0036] The signal is subjected to a discrete-time Fourier transform to obtain the frequency domain signal. The frequency domain signal is divided into s frequency bands, and an inverse Fourier transform is performed on the frequency domain signal within each frequency band to obtain the time domain signal x1(i), x2(i)…x within each frequency band. s (i).

[0037] Similar to the harmonic wavelet packet transform process described above, y(i) and z(i) are transformed to obtain y1(i), y2(i)...y s (i) and z1(i),z2(i)…z s (i).

[0038] Calculate the correlation coefficient ρ along the x-axis and y-axis for the time-domain signals of each frequency band. xy The correlation coefficient ρ between the x-axis and z-axis xz The correlation coefficient ρ between the y-axis and the z-axis yz .

[0039] The formula for calculating the correlation coefficient is as follows, to calculate the correlation coefficient ρ between the x and y axes. xy For example.

[0040]

[0041] Calculate the sum of the absolute values ​​of the correlation coefficients within each frequency band.

[0042] ρ sum =|ρ xy |+|ρ xz |+|ρ yz |

[0043] Select the σ-th frequency band with the largest sum of absolute values ​​of the correlation coefficients [m] σ ,nσ [This is used as the frequency band for calculating vibration energy.]

[0044] As a more concrete example, if the vibration signal is decomposed into 3 layers (j), the calculated ρ values ​​within each frequency band... xy ρ xz ρ yz ρ sum like Figure 2 As shown, ρ in the frequency band [1920-2240Hz] sum The maximum frequency band is therefore determined to be [1920-2240Hz] for calculating vibration energy.

[0045] Once the frequency band is determined, vibration data in the radial direction relative to the plane of rotation of the sheave is preferentially selected for calculating the vibration energy. The calculation process for vibration energy is as follows:

[0046] If the x and z axes are radial and the y axis is normal, select the x-axis frequency band [m σ ,n σ Vibration energy is calculated from the vibration data.

[0047] One method for calculating vibration energy is to select vibration data that has undergone wavelet packet transform, for example, selecting the time-domain signal x within the σ-th frequency band. σ (i) Calculate the vibration energy E for the time-domain signal x. σ (i) Summate and calculate the arithmetic square root. The formula is as follows:

[0048]

[0049] Another method for calculating vibrational energy is to perform an FFT on x(i) to obtain X(j), and then select the frequency band [m]. σ ,n σ The amplitude Y within ] σ (l) Perform FFT energy calculation

[0050] It should be noted that the method in this embodiment does not determine the wear stage of the rope wheel or set the wear warning threshold when collecting vibration data of the rope wheel for the first time. Instead, it can only determine the wear stage and set the wear warning threshold after collecting vibration data for multiple cycles.

[0051] Therefore, the total data acquisition period n for determining the wear stage should not be less than d+1, where d≥3. In this embodiment, d is set to 6, meaning that the wear stage determination is performed only after acquiring vibration data for at least 7 cycles.

[0052] An exemplary method for determining the wear stage is as follows:

[0053] Plot the vibration energy data from the nd-th cycle to the current cycle on a two-dimensional coordinate system, with the vertical axis representing vibration energy and the horizontal axis representing the time period for collecting the rope wheel vibration data. Use the least squares method to fit a univariate function curve to the (d+1)-th set of vibration energy data to obtain the slope k of the function. If k < -k thr If -k thr <k<k thr If k > k, then the wear stage is determined to be the stable wear stage; thr If k is the wear stage, then the wear stage is determined to be the severe wear stage; where k thr Given a threshold, k thr >0.

[0054] Specific examples are as follows Figure 3 As shown, point O represents the vibration energy calculated from the first collected vibration data at the start of monitoring, and point A represents the vibration energy data from the (d+1)th time. The slope k obtained by fitting the seven vibration energy data points between A and O is -0.25. A slope threshold k is set. thr The value is 0.04. k < -k thr The monitoring was initiated during the break-in and wear phase. During this phase, no wear warning threshold was set, and vibration data of the rope pulley was collected continuously at the preset cycle (30 days).

[0055] When time B is monitored, the slope k obtained by fitting the seven vibration energy data between A and B is -0.03. At this time, the -k condition is satisfied for the first time. thr <k<k thr The condition is that the vibration energy calculated based on the collected vibration data at point A' indicates that the sheave has entered a stable wear stage at time A'. Therefore, a wear warning threshold can be set at time B, and the average of the seven vibration energies between A' and B, i.e., the baseline value E, can be calculated. b The value is 16.3, and the first warning coefficient β1 is set to 3, according to the wear warning threshold E. thr =β1E b E was calculated thr The wear warning threshold is 48.9. thr Once set, it is fixed and will not be repeated later.

[0056] Then, the vibration data of the rope pulley is collected again according to the preset cycle. When the monitoring reaches time C, the slope k obtained by fitting the vibration energy data is 0.042, and at this time, k>k is satisfied for the first time. thr Under certain conditions, the system enters a stage of severe wear. When monitoring reaches time D, the vibration energy data is 50.1, exceeding the threshold, and a wear warning is issued.

[0057] The method in this embodiment can also be used to implement early warning of sheave wear on equipment such as cranes, hoists, and cable cars.

[0058] Example 2

[0059] The difference between this embodiment and Embodiment 1 is that, as Figure 4 As shown, after collecting vibration data of the rope pulley in the 7th cycle, the least squares method was used to fit a univariate function to the 7 data points from E to F, and the slope k of the function was found to be -0.08. Let k... thr It is 0.04, satisfying -k thr <k<k thr Based on the given conditions, it can be determined that the sheave was in a stable wear stage during the first cycle of monitoring.

[0060] At this point, a wear warning threshold can be set, and the average vibration energy E between EF and 7 other values ​​can be calculated. b The value is 16.1. The threshold coefficient is set to 3 based on the first warning coefficient β1, according to formula E. thr =β1E b The wear warning threshold E is calculated. thr It is 48.3.

[0061] Continue collecting vibration data from the rope pulley at the preset cycle. When the monitoring reaches time C, the slope k obtained from fitting the vibration energy data is 0.042. At this point, the first condition k > k is met. thr The condition indicates that the system has entered a severe wear stage; when monitoring reaches time D, the vibration energy is 50.1, exceeding the wear warning threshold E. thr It issues a wear and tear warning.

[0062] Example 3

[0063] The difference between this embodiment and Embodiment 1 is that, as Figure 5 As shown, after collecting vibration data of the rope pulley in the 7th cycle, the least squares method was used to fit a univariate function to the 7 data points from G to H, and the slope k of the function was found to be 0.16. Let k... thr The value is 0.04, satisfying k>k thr Based on the given conditions, it can be determined that the rope wheel was in a stage of severe wear during the first cycle of monitoring.

[0064] At this point, a wear warning threshold can be set, corresponding to a baseline value E for the severe wear stage. b Compared with the set baseline value E during the stable wear phase b The difference is that the baseline value E is different here. b This is the vibration energy value for the first cycle. The vibration energy at point G is 16.4, and the baseline value E is set. b It is 16.4.

[0065] The second early warning coefficient β2 is set to 2.5, according to formula E. thr =β2E b The wear warning threshold E is calculated. thr The vibration energy at point H is 55, exceeding the wear warning threshold E. thr It issues a wear and tear warning.

[0066] Example 4

[0067] The difference between this embodiment and Embodiment 1 is that an example of determining the frequency band for calculating vibration energy using a biaxial vibration sensor is given; the rest of the content is the same as in Embodiment 1.

[0068] For a biaxial vibration sensor, harmonic wavelet packet transform is performed on the vibration data x(i) and y(i) in the x and y directions.

[0069] Taking x(i) as an example, in harmonic wavelet packet transform, the highest analysis frequency of the signal is The vibration data signal is decomposed into j (j = 0, 1, 2, 3, ...) frequency bands, and the bandwidth B of each band can be expressed as B = 2. -j f h s is the number of frequency bands, s = 2 j The upper limit m and lower limit n of each analysis frequency band are respectively

[0070] The signal is subjected to a discrete-time Fourier transform to obtain the frequency domain signal. The frequency domain signal is divided into s frequency bands, and an inverse Fourier transform is performed on the frequency domain signal within each frequency band to obtain the time domain signal x1(i), x2(i)…x within each frequency band. s (i). Similar to the harmonic wavelet packet transform process described above, y(i) is transformed to obtain y1(i), y2(i)...y s (i).

[0071] For the x and y axis time-domain signals within each frequency band, calculate the absolute value of the correlation coefficient |ρ| along the x and y axes. xy |. Select the σ-th frequency band with the largest absolute value of the correlation coefficient [m] σ ,n σ [This is used as the frequency band for calculating vibration energy.]

[0072] In a preferred embodiment, the vibration data signal is decomposed into 3 layers (j), and the calculated |ρ| within each frequency band is... xy |as Figure 6 As shown, |ρ| in the range of [1600-1920Hz] xy The maximum frequency band for calculating vibration energy is [1600-1920Hz].

[0073] The present invention has been described in detail above through specific embodiments and examples, but these are not intended to limit the invention. Many modifications and improvements can be made by those skilled in the art without departing from the principles of the invention, and these should also be considered within the scope of protection of the present invention.

Claims

1. A method for early warning of rope and pulley wear based on vibration signals, characterized in that, Vibration data of the rope wheel is collected according to a preset cycle, and its vibration energy is calculated based on the vibration data collected in each cycle. When the total collection cycle n is not less than d+1, the wear stage of the rope wheel in the nd cycle is determined based on the vibration energy, where d≥3. The wear stage is divided into running-in wear stage, stable wear stage and severe wear stage. If no wear warning threshold is set, when the wear stage of the pulley is a stable wear stage or a severe wear stage in the nd cycle, a wear warning threshold is set; it is determined whether the vibration energy of the current cycle exceeds the wear warning threshold, and if it does, a wear warning signal is issued. The method for setting the wear warning threshold is as follows: When the sheave is in the stable wear stage during the nd cycle, E thr =β1E b E thr β1 is the first warning coefficient, and E is the wear warning threshold. b Baseline value; When the sheave is in the severe wear stage during the nd cycle, E thr =β2E b E thr β2 is the wear warning threshold, and E is the second warning coefficient. b Baseline value; When the wear stage of the sheave is the stable wear stage in the nd cycle, the baseline value E b The average vibrational energy from the nd-th cycle to the current cycle; When the wear stage of the sheave is severe wear stage during the nd cycle, the baseline value E b Let be the vibrational energy of the nd period.

2. The method for early warning of rope and pulley wear based on vibration signals according to claim 1, characterized in that, The method for determining the wear stage of the sheave in the nd cycle based on the vibration energy is as follows: Plot the vibration energy data from the nd-th cycle to the current cycle on a two-dimensional coordinate system, with the vertical axis representing vibration energy and the horizontal axis representing the time period for collecting the rope wheel vibration data. Use the least squares method to fit a univariate function curve to the (d+1)-th set of vibration energy data to obtain the slope k of the function. If k < -k thr If -k thr <k< k thr If k > k, then the wear stage is determined to be the stable wear stage; thr If k is the wear stage, then the wear stage is determined to be the severe wear stage; where k thr Given a threshold, k thr >0.

3. The method for early warning of rope and pulley wear based on vibration signals according to claim 1, characterized in that, The first warning coefficient β1 is greater than the second warning coefficient β2.

4. The method for early warning of rope and pulley wear based on vibration signals according to claim 1, characterized in that, The vibration data of the rope pulley is collected every 30 days as the preset cycle.

5. The method for early warning of rope and pulley wear based on vibration signals according to claim 1, characterized in that, The sheave is at least one of the following in the elevator system: traction sheave, guide sheave, anti-rope sheave, or compensating sheave.

6. The method for early warning of rope and pulley wear based on vibration signals according to claim 5, characterized in that, The vibration data collected from the sheaves are the vibration data of the sheaves during the section of the elevator running at a constant speed from the end floor to the end floor.

7. The method for early warning of rope and pulley wear based on vibration signals according to claim 1, characterized in that, The method for calculating vibration energy based on the vibration data is as follows: determine the frequency band for calculating vibration energy, sum the time-domain signals of the vibration data in a single direction within the frequency band after processing, and calculate the vibration energy by performing an arithmetic square root.

8. The method for early warning of rope and pulley wear based on vibration signals according to claim 7, characterized in that, The single direction is the radial direction relative to the plane of rotation of the sheave.

9. The method for early warning of rope and pulley wear based on vibration signals according to claim 7, characterized in that, The method for determining the frequency band for calculating vibration energy is as follows: obtain the time-domain signal of each axis in each frequency band of vibration data through harmonic wavelet packet transform, calculate the sum of the absolute values ​​of the correlation coefficients between each axis based on the time-domain signal of each axis in each frequency band, and determine the frequency band with the largest sum of the absolute values ​​of the correlation coefficients as the frequency band for calculating vibration energy.

Images