Steel rail vibration signal noise reduction method based on ultra-weak fiber grating technology

Abstract

The invention discloses a rail vibration signal noise reduction method based on an ultra-weak fiber bragg grating technology. The method comprises the following steps: step 1, detecting a rail vibration signal by using the ultra-weak fiber bragg grating technology; and step 2, carrying out noise reduction processing on the detected steel rail vibration signal by adopting a wavelet noise reduction method. Wherein the wavelet noise reduction method comprises the step of selecting a wavelet basis function, a threshold function and a threshold rule, the selected wavelet basis function has tight supporting performance, large disappearance moment and few extreme value points, the threshold function comprises a hard threshold function and a soft threshold function, and the threshold rule comprises medium unbiased risk estimation, a fixed threshold, a heuristic threshold and an extreme value threshold.

Description

Technical Field

[0001] The present invention relates to the technical field of rail detection, and in particular to a rail vibration signal acquisition and noise reduction method based on ultra-weak fiber grating technology. Background Art

[0002] As a kind of optical fiber sensing method, ultra-weak fiber Bragg grating cable sensing technology has the advantages of being free from electromagnetic interference, strong resistance to environmental interference, long service life, and can be applied to large-scale scene detection. However, at the same time, while collecting data, it is inevitable that surrounding noise (the sound of machine operation, electromagnetic interference signal, etc.) will be mixed in the original pure vibration signal. The noise component will distort the signal, thereby interfering with the extraction of rail vibration signal characteristics in the railway line. Therefore, the noise reduction of the rail vibration signal is of great significance for the subsequent classification of rail vibration signal events. Therefore, it is necessary to perform noise reduction processing on the collected vibration signal. The better the noise reduction effect, the closer the signal obtained after noise reduction is to the original vibration signal of the rail, and the closer the extracted eigenvalue is to the real vibration eigenvalue. Therefore, the quality of the noise reduction effect will affect the accuracy of subsequent classification.

[0003] In traditional signal denoising methods, Fourier transform is usually used to divide the signal into high-frequency and low-frequency components, and the noise is removed by removing the high-frequency part. This method will eliminate some useful information while eliminating high-frequency noise, causing distortion in the high-frequency part of the signal. Therefore, it often cannot achieve good results when processing non-stationary and highly mutant signals. The vibration signal of the rail is nonlinear and non-steady. Therefore, the traditional fast Fourier transform (FFT) cannot accurately reflect the detailed characteristics of the signal, and it is necessary to select a signal denoising processing method that is more suitable for the signal. Summary of the invention

[0004] Wavelet transform is a method of localization analysis in the time and frequency domains. Its window size is fixed, but its shape can be changed. Therefore, it can overcome the limitations of Fourier transform in the time domain and realize local analysis of time and frequency. Wavelet transform is suitable for processing non-stationary and sudden change signals. It has higher frequency resolution and lower time resolution in the low-frequency part, and higher time resolution and lower frequency resolution in the high-frequency part, so it can focus on any details of the input signal. Therefore, wavelet transform (WT) is suitable for analyzing the changes of non-stationary signals.

[0005] The present invention aims to provide a method for reducing noise of rail vibration signals based on ultra-weak fiber Bragg grating cables, which adopts wavelet denoising to make the collected rail vibration signals closer to pure vibration signals, thus facilitating data analysis.

[0006] More specifically, the present invention proposes a method for reducing noise of rail vibration signals based on ultra-weak fiber Bragg grating technology, comprising the following steps:

[0007] Step 1: Detect rail vibration signals using ultra-weak fiber Bragg grating technology; and

[0008] Step 2: Using the wavelet noise reduction method to perform noise reduction processing on the detected rail vibration signal;

[0009] Among them, the wavelet denoising method includes selecting wavelet basis functions, threshold functions and threshold rules, wherein the selected wavelet basis functions have compact support, large vanishing moments and fewer extreme value points, the threshold functions include hard threshold functions and soft threshold functions, and the threshold rules include unbiased risk estimation, fixed thresholds, heuristic thresholds and extreme value thresholds.

[0010] According to an embodiment of the present invention, the method further comprises step 3 of evaluating the noise reduction effect.

[0011] According to an embodiment of the present invention, the noise reduction effect is evaluated using a signal-to-noise ratio (SNR) and a root mean square error (RMSE).

[0012] According to an embodiment of the present invention, the wavelet basis function is selected from Db2, Db4, Db6, Db8, Sym2, Sym4, Sym6 and Sym8.

[0013] According to an embodiment of the present invention, the wavelet basis function is selected from Db8 and Sym8, the threshold function is a soft threshold function, and the threshold rule is a fixed threshold.

[0014] According to an implementation scheme of the present invention, in step 2, the wavelet denoising method performs wavelet decomposition on the detected rail vibration signal, sets the high-frequency coefficients representing the noise signal obtained in the wavelet decomposition to zero, and reconstructs the signal using the zeroed high-frequency coefficients, the non-zeroed high-frequency coefficients and the retained low-frequency coefficients, thereby removing the high-frequency noise in the signal.

[0015] According to an embodiment of the present invention, the noise reduction effect evaluation includes calculating the SNR and RMSE of the signal after performing signal noise reduction on the wavelet reconstructed signal by using the noise reduction function in the wavelet decomposition. If the SNR is larger and the RMSE is smaller, it indicates that the noise reduction effect is better.

[0016] The embodiment of the present invention also provides a rail vibration signal noise reduction device based on ultra-weak fiber Bragg grating technology, comprising:

[0017] Vibration detection module, using ultra-weak fiber Bragg grating technology to detect rail vibration signals;

[0018] The data processing and noise reduction module uses the wavelet noise reduction method to perform noise reduction on the detected rail vibration signal; and the noise reduction effect evaluation module is used to evaluate the noise reduction effect.

[0019] An embodiment of the present invention further provides an electronic device, comprising: a memory and one or more processors;

[0020] The memory is used to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flow chart of a rail vibration signal noise reduction method based on ultra-weak fiber Bragg grating technology according to an embodiment of the present invention;

[0022] Figure 2 2 is a schematic diagram of the structure of a rail vibration signal noise reduction device based on ultra-weak fiber Bragg grating technology according to an embodiment of the present invention;

[0023] Figure 3 A schematic diagram of the structure of an electronic device according to an embodiment of the present invention;

[0024] Figure 4 It is a noise reduction effect diagram of different wavelet basis functions according to the implementation scheme of the present invention;

[0025] Figure 5 This is a comparison diagram of the original vibration signal and the signal after noise reduction when the rail vibration signal noise reduction method based on ultra-weak fiber Bragg grating technology according to an embodiment of the present invention is applied to a fastener failure condition. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the embodiments described are only part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0027] The features and performance of the present invention are further described in detail below in conjunction with the embodiments:

[0028] Figure 1 This is a flow chart of a rail vibration signal noise reduction method based on ultra-weak fiber Bragg grating technology according to an embodiment of the present invention; as shown in the figure, the rail vibration signal noise reduction method based on ultra-weak fiber Bragg grating technology may include the following steps:

[0029] First, ultra-weak fiber Bragg grating technology can be used to detect rail vibration signals. For example, a vibration detection device based on ultra-weak fiber Bragg grating optical cables can be used for detection, such as a distributed optical fiber acoustic wave / vibration sensing system (DAS). The vibration detection device may include an ultra-weak fiber Bragg grating optical cable, which can emit and transmit modulated optical signals, and obtain a reflected optical signal after the modulated optical signal is reflected by the optical cable, obtain a vibration signal to be measured based on the reflected optical signal, and transmit the vibration signal to be measured to a data processing center for further processing.

[0030] After obtaining the vibration signal of the rail, for example, the wavelet denoising method is used to perform denoising on the rail vibration signal, so that the obtained vibration signal is as close to the original vibration signal of the rail as possible and the interference of the surrounding environmental noise is reduced. The wavelet denoising method may include selecting a wavelet basis function, a threshold function, and a threshold rule.

[0031] The wavelet basis function selects a wavelet function with compact support, a wavelet function with a large vanishing moment and a small number of extreme value points. Compact support means that the wavelet function should have a limited non-zero area in the time domain to ensure that the wavelet transform has good localization characteristics; the larger the vanishing moment of the wavelet basis function, the more effective it is to remove the noise in the low-frequency part; reducing the number of extreme value points can effectively reduce the interference of high-frequency noise. For example, the wavelet basis function can be selected from the wavelet basis function selected from Db2, Db4, Db6, Db8, Sym2, Sym4, Sym6, Sym8 and Haar, etc., preferably Db8 and Sym8. The threshold function includes a hard threshold function and a soft threshold function, and the threshold rule includes a neutral unbiased risk estimate, a fixed threshold, a heuristic threshold and an extreme value threshold. For example, the threshold function is preferably a soft threshold function, and the threshold rule is preferably a fixed threshold.

[0032] More specifically, for example, the collected rail vibration signal can be subjected to a 3-layer db4 wavelet decomposition, and the high-frequency coefficients representing the noise signal obtained in the wavelet decomposition can be set to zero; then, the zeroed high-frequency coefficients, the non-zeroed high-frequency coefficients and the retained low-frequency coefficients are used for signal reconstruction, thereby removing the high-frequency noise in the signal. It should be understood that these wavelet basis functions, threshold functions and threshold rules are known in the art and are not described in detail here.

[0033] Finally, the noise reduction effect is evaluated. For example, the noise reduction effect can be evaluated using the signal-to-noise ratio (SNR) and the root mean square error (RMSE).

[0034] In digital signal processing, the quantitative evaluation index of filtering effect can be signal-to-noise ratio (SNR) and root mean square error (RMSE). The signal-to-noise ratio refers to the ratio between the strength of useful information in the signal and the strength of noise. The calculation formula of the signal-to-noise ratio is:

[0035]

[0036] The physical meaning of the root mean square is the average of the sum of the squares of the difference between the filtered signal and the original signal, which is used to evaluate the difference between the noise-reduced signal and the original signal. The calculation formula for the root mean square is:

[0037]

[0038] According to the above formula, the larger the SNR, the more effective information is contained in the denoised signal, and the smaller the RMSE, the more information of the original signal is retained in the denoised signal. Therefore, the evaluation criteria for the denoising effect can be defined: for the noisy signal s(t), after denoising the signal using different denoising methods, the SNR and RMSE of the denoised signal f(t) are calculated. If the SNR is larger and the RMSE is smaller, the denoising effect is better.

[0039] Figure 2 Schematic diagram of the structure of a rail vibration signal noise reduction device based on ultra-weak fiber Bragg grating technology according to an embodiment of the present invention. Figure 2 As shown, the device includes: a vibration detection module 210, which uses ultra-weak fiber Bragg grating technology to detect rail vibration signals; a data processing and noise reduction module 220, which uses wavelet noise reduction method to perform noise reduction processing on the detected rail vibration signals; and a noise reduction effect evaluation module 230, which is used to evaluate the noise reduction effect.

[0040] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, the electronic device includes a processor 310, a memory 320, an input device 330 and an output device 340; the number of the processor 310 in the electronic device can be one or more. Figure 3 A processor 310 is taken as an example; the processor 310, the memory 320, the input device 330 and the output device 340 in the electronic device can be connected via a bus or other means. Figure 3 The example of connecting through bus is taken in the following.

[0041] The memory 320, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the rail vibration signal noise reduction method based on ultra-weak fiber Bragg grating technology in the embodiment of the present invention (for example, vibration detection module 210; data processing and noise reduction module 220; noise reduction effect evaluation module 230). The processor 310 executes various functional applications and data processing of the electronic device by running the software programs, instructions and modules stored in the memory 320, that is, realizing the above-mentioned rail vibration signal noise reduction method based on ultra-weak fiber Bragg grating technology.

[0042] The memory 320 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 320 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some instances, the memory 320 may further include a memory remotely arranged relative to the processor 310, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The input device 330 may be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the electronic device. The output device 340 may include a display device such as a display screen.

[0043] In order to further verify the method of the present invention, a subway rail section is taken as the research object, and the method of the present invention is used to conduct experiments. That is, in order to study the selection of threshold function when the wavelet denoising method is used for rail vibration signals, a group of vibration signals under typical working conditions are selected, and several common wavelet basis functions are selected to conduct noise reduction effect comparison experiments under the unbiased threshold rules of soft threshold and hard threshold respectively:

[0044] Detection hardware: Distributed fiber acoustic / vibration sensing system (DAS);

[0045] Wavelet basis functions: Db2, Db4, Db6, Db8, Sym2, Sym4, Sym6, Sym8 and Haar;

[0046] Threshold function: hard threshold function and soft threshold function;

[0047] Threshold rules: unbiased risk estimation, fixed threshold, heuristic threshold and extreme threshold;

[0048] Effect evaluation indicators: signal-to-noise ratio (SNR) and root mean square error (RMSE);

[0049] The specific experimental results are as follows:

[0050] Table 1: Noise reduction effect of different wavelet basis functions on hard threshold function

[0051]

[0052] Table 2: Noise reduction effect of different wavelet basis functions in soft threshold function

[0053]

[0054] By comparing Table 1 and Table 2, it can be seen that when several different wavelet basis functions are selected, the signal-to-noise ratio when using the soft threshold function for denoising is greater than the signal-to-noise ratio of the hard threshold function for denoising, and the root mean square RMSE is smaller than the root mean square RMSE of the hard threshold function for denoising. The effects of several wavelet basis functions under the soft threshold function for denoising are better than those under the hard threshold function for denoising. Therefore, the soft threshold function is preferred for denoising in wavelet function denoising.

[0055] On the basis of soft threshold function denoising, the denoising effects of the above wavelet basis functions under different threshold rules are compared, and the optimal threshold rule is selected. The results are shown in the following table:

[0056] Table 3: De-noising effect of db2 wavelet basis function matching under four threshold rules

[0057]

[0058] Table 4: De-noising effect of db4 wavelet basis function matching four threshold rules

[0059]

[0060] Table 5: De-noising effect of db6 wavelet basis function matching four threshold rules

[0061]

[0062] Table 6: De-noising effect of db8 wavelet basis function matching four threshold rules

[0063]

[0064] Table 7: Denoising effect of sym2 wavelet basis function matching four threshold rules

[0065]

[0066] Table 8: De-noising effect of sym4 wavelet basis function matching four threshold rules

[0067]

[0068] Table 9: De-noising effect of sym6 wavelet basis function matching four threshold rules

[0069]

[0070] Table 10: De-noising effect of sym8 wavelet basis function matching four threshold rules

[0071]

[0072] Table 11: Denoising effect under four threshold rules matching Haar wavelet basis function

[0073]

[0074] Table 12: Selection results of the best basis function and the best threshold rule

[0075]

[0076] Table 12 summarizes the experimental results of the best wavelet basis function and the best threshold rule in Table 3-11. From the comparison of the signal-to-noise ratio and the root mean square value in Table 12 above, it can be concluded that for the noisy signal of the rail vibration disease monitoring experiment, the best effect can be achieved when the db8 wavelet basis function under the Daubechies wavelet system is used for denoising with a fixed threshold and a soft threshold function. Figure 4 This is a diagram of denoising effects of different wavelet basis functions according to an embodiment of the present invention, showing the final denoising effects of different signal denoising methods. From this diagram, it can be seen that the best denoising effect can be achieved by using the db8 wavelet denoising method.

[0077] Figure 5 The figure is a comparison of the original vibration signal and the signal after noise reduction applied to the rail vibration signal noise reduction method based on ultra-weak fiber Bragg grating technology under the fastener failure condition according to the embodiment of the present invention. That is, we intercepted the event window of 0.2 seconds of vibration, and performed wavelet (using the db8 wavelet basis function under the Daubechies wavelet system, fixed threshold, and soft threshold function) decomposition and reconstruction on the rail vibration signal under the fastener failure to obtain the denoised signal.

[0078] pass Figure 5 From the comparison, we can see that the signal after wavelet denoising is smoother than the original vibration signal, and a large number of "burrs" have been filtered out, which proves that most of the high-frequency noise has been filtered out through denoising, and the low-frequency vibration signal has been retained. On this basis, many vibration characteristics of the original signal have been retained. The denoising effect is good and it is suitable for the next step of signal feature extraction.

[0079] The above description of the embodiments is to facilitate the understanding and application of the present invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the embodiments herein, and improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the scope of protection of the present invention.

Claims

1. A method for reducing noise of rail vibration signals based on ultra-weak fiber Bragg grating technology, characterized in that: The following steps are involved: Step 1: Detect rail vibration signals using ultra-weak fiber Bragg grating technology; as well as Step 2: Using wavelet noise reduction method to perform noise reduction processing on the detected rail vibration signal; Among them, the wavelet denoising method includes selecting wavelet basis functions, threshold functions and threshold rules, wherein the selected wavelet basis functions have compact support, large vanishing moments and fewer extreme value points, the threshold functions include hard threshold functions and soft threshold functions, and the threshold rules include unbiased risk estimation, fixed thresholds, heuristic thresholds and extreme value thresholds.

2. The method according to claim 1, characterized in that The method further includes step 3 of evaluating the noise reduction effect.

3. The method according to claim 2, characterized in that The signal-to-noise ratio (SNR) and root mean square error (RMSE) were used to evaluate the noise reduction effect.

4. The method according to claim 1, characterized in that: The wavelet basis functions are selected from Db2, Db4, Db6, Db8, Sym2, Sym4, Sym6 and Sym8.

5. The method according to claim 1, characterized in that: The wavelet basis functions are selected from Db8 and Sym8, the threshold function is a soft threshold function, and the threshold rule is a fixed threshold.

6. The method according to claim 1 or 2, characterized in that: In step 2, the wavelet denoising method performs wavelet decomposition on the detected rail vibration signal, sets the high-frequency coefficients representing the noise signal obtained in the wavelet decomposition to zero, and reconstructs the zeroed high-frequency coefficients, the non-zeroed high-frequency coefficients and the retained low-frequency coefficients to remove the high-frequency noise in the signal.

7. The method according to claim 6, characterized in that The evaluation of the noise reduction effect includes calculating the SNR and RMSE of the signal after performing noise reduction on the wavelet reconstructed signal using the noise reduction function in the wavelet decomposition. If the SNR is larger and the RMSE is smaller, the noise reduction effect is better.

8. A rail vibration signal noise reduction device based on ultra-weak fiber Bragg grating technology, characterized in that: include: Vibration detection module, using ultra-weak fiber Bragg grating technology to detect rail vibration signals; The data processing and noise reduction module uses the wavelet noise reduction method to reduce the noise of the detected rail vibration signal; and a noise reduction effect evaluation module for evaluating the noise reduction effect.

9. An electronic device, characterized in that: include: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

Images