Steel rail breakage detection method and system
By using 13-bit Barker code encoding and an adaptive matched filter for track environment to process guided wave signals in the railway rail inspection system, the problem of environmental interference in long-distance inspection is solved, enabling effective inspection at greater distances and reducing costs.
Patent Information
- Application Number
- CN202411168654.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-22
AI Technical Summary
Existing ultrasonic guided wave-based railway rail breakage detection systems are greatly affected by environmental factors during long-distance detection, making it difficult to achieve effective detection, and the equipment layout and maintenance costs are high.
The excitation signal is encoded using 13-bit Barker code, and the guided wave signal is processed by an adaptive matched filter for the track environment to improve the signal-to-noise ratio, eliminate environmental interference, and achieve long-distance detection.
It improves the signal-to-noise ratio of guided wave signals, extends the detection distance, reduces labor and equipment costs, and effectively reduces the adverse effects of the environment on detection.
Smart Images

Figure CN118991872A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of non-destructive detection technology, and in particular, relates to a method and system for detecting broken rails. Background Art
[0002] At present, the main methods for detecting broken rails on railways include the detection system of broken rails on railways based on ultrasonic guided waves. Usually, the maximum effective detection distance of a set of broken rail detection systems on railways based on ultrasonic guided waves is about 2km. Taking conventional railways as an example, the distance between two adjacent sections is about 10km. Therefore, it is necessary to arrange multiple sets of equipment in the section. The economic and manpower investment in the early installation and later maintenance is large, which makes it difficult to meet the needs of long-distance seamless rail broken inspection in railway sections.
[0003] In order to realize long-distance detection of railway rail breakage based on ultrasonic guided waves, it is particularly important to improve the energy of the excitation signal emitted by the excitation end and the signal-to-noise ratio of the waveguide signal received by the receiving end. The excitation signal sent by the excitation end is transmitted from the electrical signal, converted from the transducer to the acoustic signal, and coupled to the rail. After long-distance transmission, the waveguide signal received by the receiving end has a low correlation with the excitation signal; and the waveguide signal is greatly affected by the rail auxiliary structure, the roadbed and the external environment (temperature, humidity, etc.). The original excitation signal is flipped in the time domain as a matched filter, and the received waveguide signal is matched and filtered. The signal-to-noise ratio of the waveguide signal after matched filtering is improved, which cannot effectively realize the long-distance detection of railway rail breakage, and the influence of the railway environment on the detection of railway rail breakage cannot be eliminated.
[0004] Therefore, there is an urgent need for a method and system that will not be affected by the railway environment and can achieve long-distance detection of broken railway rails. Summary of the invention
[0005] To solve the above problems, the present invention provides a rail break detection method and system, which adopts 13-bit Barker code to encode the excitation signal, improves the energy and average power carried by the excitation signal, and is more suitable for long-distance rail break detection; the excitation signal is converted from an electrical signal to an acoustic wave, and coupled to the rail, so that the acoustic wave is transmitted on the rail, and a track environment adaptive matching filter is determined according to the waveguide signal received by the sensor closer to the acoustic wave coupling position, and the waveguide signal received by the sensor farther from the acoustic wave coupling position is matched and filtered by the track environment adaptive matching filter, so that a rail detection signal with no interference from the track environment can be obtained, that is, a rail detection signal with no interference from the track environment can be obtained, which is used for rail break detection.
[0006] The present invention is achieved through the following technical solutions:
[0007] In a first aspect, an embodiment of the present disclosure provides a method for detecting broken rails, the method comprising:
[0008] Generate an original excitation signal, and perform phase encoding on the original excitation signal to obtain an encoded excitation signal;
[0009] Converting the coded excitation signal from an electrical signal into an acoustic wave, and coupling it to the rail, so that the acoustic wave propagates along the rail;
[0010] receiving an acoustic wave at a first position, generating a first waveguide signal, preprocessing the first waveguide signal, and determining a track environment adaptive matched filter according to the preprocessed first waveguide signal;
[0011] Receive the sound wave at the second position, generate a second waveguide signal, use the track environment adaptive matched filter to perform matched filtering on the second waveguide signal to obtain a rail detection signal, and detect whether the rail is broken based on the rail detection signal.
[0012] Furthermore,
[0013] Generate a 5-period sinusoidal signal with a sampling frequency of 200 kHz and a center frequency of a preset frequency, which is used as the original excitation signal;
[0014] The original excitation signal is phase-encoded using a Barker code to obtain the coded excitation signal.
[0015] Furthermore,
[0016] The coded excitation signal is power-amplified, and the power-amplified coded excitation signal is converted from an electrical signal to an acoustic wave, which is coupled to a steel rail so that the acoustic wave propagates along the steel rail.
[0017] Furthermore,
[0018] receiving the sound wave at the first position, generating the first waveguide signal, and performing normalization processing on the first waveguide signal;
[0019] The first waveguide signal after normalization is flipped in time domain to determine the track environment adaptive matching filter.
[0020] Furthermore,
[0021] receiving the sound wave at the second position, generating the second waveguide signal, and performing time domain convolution on the track environment adaptive matched filter and the second waveguide signal to obtain the rail detection signal;
[0022] According to a preset algorithm or a preset detection process, the rail detection signal is used to detect whether the rail is broken.
[0023] Furthermore,
[0024] Use the Hanning window modulation method to generate a 5-cycle sine signal.
[0025] Furthermore,
[0026] The original excitation signal is phase-encoded using a 13-bit Barker code to obtain the coded excitation signal.
[0027] Furthermore,
[0028] The noise signal of the track environment is collected and used to determine the signal-to-noise ratio with the rail detection signal.
[0029] In a second aspect, an embodiment of the present disclosure further provides a rail break detection system, the system comprising:
[0030] A signal generating module, a signal coupling module, a first signal processing module and a second signal processing module;
[0031] The signal generating module is used to generate an original excitation signal, and perform phase encoding on the original excitation signal to obtain an encoded excitation signal;
[0032] The signal coupling module is used to convert the coded excitation signal from an electrical signal into an acoustic wave, and couple it to the rail so that the acoustic wave propagates along the rail;
[0033] The first signal processing module is used to receive the sound wave at the first position, generate a first waveguide signal, preprocess the first waveguide signal, and determine the track environment adaptive matched filter according to the preprocessed first waveguide signal;
[0034] The second signal processing module is used to receive the sound wave at the second position, generate a second waveguide signal, use the track environment adaptive matched filter to perform matched filtering on the second waveguide signal to obtain a rail detection signal, and detect whether the rail is broken based on the rail detection signal.
[0035] Further, the signal coupling module includes: a transducer and a power amplifier;
[0036] The power amplifier is used to amplify the power of the coded excitation signal;
[0037] The transducer is used to convert the power-amplified coded excitation signal from an electrical signal into an acoustic wave, and couple it to the rail.
[0038] Furthermore, the first signal processing module includes: a first sensor, a first acquisition instrument and a first computing unit;
[0039] The first sensor is used to receive the sound wave at the first position and generate the first guided wave signal;
[0040] The first acquisition instrument is used to collect the first waveguide signal of the first sensor and send it to the first computing unit;
[0041] The first operation unit is used to normalize the first waveguide signal, perform time domain flipping on the normalized first waveguide signal, and determine the track environment adaptive matching filter.
[0042] Furthermore, the second signal processing module includes a second sensor, a second acquisition instrument and a second computing unit;
[0043] The second sensor is used to receive the sound wave at the second position and generate the second guided wave signal;
[0044] The second collector is used to collect the second waveguide signal of the second sensor and send it to the second computing unit;
[0045] The second operation unit is used to perform time domain convolution on the track environment adaptive matched filter and the second waveguide signal to obtain the rail detection signal, and use the rail detection signal to detect whether the rail is broken according to a preset algorithm or a preset detection process.
[0046] Compared with the prior art, the present invention has the following advantages:
[0047] 1. The signal-to-noise ratio of the waveguide signal used for rail break detection is improved, which effectively increases the distance of rail detection and reduces labor costs and equipment costs.
[0048] 2. Use the track environment adaptive filter to perform matched filtering on the waveguide signal used for rail break detection, effectively reducing the adverse effects of the track environment on the use of waveguide signals for rail break detection.
[0049] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0051] Figure 1A flow chart of a rail break detection method provided in an embodiment of the present disclosure;
[0052] Figure 2 A model structure diagram of a rail break detection method provided in an embodiment of the present disclosure;
[0053] Figure 3 An original excitation signal waveform diagram provided in an embodiment of the present disclosure;
[0054] Figure 4 A schematic diagram of a 13-bit Barker code provided in an embodiment of the present disclosure;
[0055] Figure 5 A waveform diagram of a coding excitation signal provided in an embodiment of the present disclosure;
[0056] Figure 6 A schematic diagram of a received waveguide signal waveform provided in an embodiment of the present disclosure;
[0057] Figure 7 A schematic diagram of another received waveguide signal waveform provided in an embodiment of the present disclosure;
[0058] Figure 8 A rail detection signal waveform diagram provided by an embodiment of the present disclosure;
[0059] Fig. 9 A noise signal waveform diagram of a track environment provided by an embodiment of the present disclosure;
[0060] Fig.10 A waveform diagram of a second waveguide signal of a conventional matched filter provided in an embodiment of the present disclosure;
[0061] Fig.11 A scatter plot of a signal-to-noise ratio is provided in an embodiment of the present disclosure. DETAILED DESCRIPTION
[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.
[0063] Figure 1 A flow chart of a rail break detection method provided in an embodiment of the present disclosure is as follows: Figure 1 As shown, a rail break detection method provided by an embodiment of the present disclosure includes:
[0064] S1: Generate an original excitation signal, and perform phase encoding on the original excitation signal to obtain a coded excitation signal.
[0065] Specifically, the signal generation module generates a 5-cycle sinusoidal signal with a sampling frequency fs of 200kHz and a center frequency fc of a preset frequency, which is used as the original excitation signal sig(n); the original excitation signal sig(n) is phase-encoded using a Barker code to obtain a coded excitation signal sig e (n). When selecting the preset frequency, the center frequency is preferably a frequency with a lower attenuation coefficient on the rail, which is more suitable for long-distance rail detection. The specific selection of the preset frequency can be adjusted by those skilled in the art according to the actual situation of the rail and track environment and the distance of rail detection, and no further specific limitation is made here.
[0066] S2: Convert the coded excitation signal from an electrical signal into an acoustic wave and couple it to the rail so that the acoustic wave propagates along the rail.
[0067] Specifically, the signal coupling module is used to encode the excitation signal sig e (n) The electrical signal is converted into an acoustic wave, and the acoustic wave is coupled to the rail, and the acoustic wave propagates along the rail. A first signal processing module and a second signal processing module are sequentially arranged in the direction of the acoustic wave propagation, the first signal processing module includes a first sensor, and the second signal processing module includes a second sensor, and the first sensor and the second sensor are used to receive the acoustic wave coupled to the rail. The first sensor is located at a first position, which is close to the position where the signal coupling module couples the acoustic wave, and is used to receive the acoustic wave with a relatively short propagation distance, and generate a first waveguide signal s 1_e (n); The second sensor is located at a second position, which is far away from the position where the signal coupling module couples the sound wave, and is used to receive the sound wave with a relatively long propagation distance and generate a second waveguide signal s 2_e (n).
[0068] It should be understood that the sound wave coupled to the rail by the signal coupling module continues to propagate toward the first sensor and the second sensor, and the sound wave received by the first sensor and the second sensor may be the same sound wave. After the first sensor receives the sound wave coupled to the rail by the signal coupling module, the sound wave may continue to propagate toward the second sensor. Of course, the signal coupling module may couple the sound wave converted from the same coded excitation signal to the rail at two intervals, and the first sensor and the second sensor receive two sound waves in turn.
[0069] S3: receiving the sound wave at the first position, generating a first waveguide signal, preprocessing the first waveguide signal, and determining a track environment adaptive matching filter according to the preprocessed first waveguide signal.
[0070] Specifically, the first sensor of the first signal processing module receives the sound wave at the first position and converts it into a first guided wave signal s 1_e (n), according to the first waveguide signal s 1_e (n), determine the orbit environment adaptive matched filter h e_mf (n). After the first sensor receives the sound wave, it does not affect the sound wave from continuing to propagate in the direction of the second signal processing module.
[0071] S4: receiving the sound wave at the second position, generating a second waveguide signal, performing matched filtering on the second waveguide signal using a track environment adaptive matched filter to obtain a rail detection signal, and detecting whether the rail is broken according to the rail detection signal.
[0072] Specifically, the second sensor of the second signal processing module receives the sound wave at the second position and converts it into a second guided wave signal s 2_e (n), the second waveguide signal s 2_e (n) and orbit environment adaptive matched filter h e_mf (n) Perform matched filtering and the result is used as the rail detection signal y e_mf (n). The rail detection signal y e_mf (n) In the second signal processing module, a preset algorithm is used to detect whether the rail is broken; or the rail detection signal y e_mf (n) Execute the preset detection process; for example, the rail detection signal y e_mf (n) Perform signal filtering and determine the signal amplitude after signal filtering. When the signal amplitude is lower than a preset threshold, it is considered that the rail is broken. The sound wave received by the second sensor may be the same as the sound wave received by the first sensor, but the transmission distance is longer.
[0073] It should be understood that the first position and the second position need to be set according to the specific track environment and the distance between the two sections of the railway. For example: when the distance between the two sections of the railway is greater than 5km, the broken rail is detected, and the distance between the first position and the position where the signal coupling module couples the sound wave does not exceed 20% of the total length; that is, when the distance between the two sections of the railway is 5km, the distance between the first sensor and the position where the signal coupling module couples the sound wave is not more than 1km; the second position can be within the range of the two sections of the railway, at the other end opposite to the signal coupling module, that is, when the distance between the two sections of the railway is 5km, the second sensor is located 5km away from the position where the signal coupling module couples the sound wave. In the disclosed embodiment, the shorter distance and the longer distance are not specifically limited, but are used to illustrate the positional relationship between the first sensor and the second sensor and the transducer (sound wave coupling position).
[0074] In some examples, generating an original excitation signal and performing phase encoding on the original excitation signal to obtain an encoded excitation signal specifically includes:
[0075] S11: Generate a 5-cycle sinusoidal signal with a sampling frequency of 200 kHz and a center frequency of a preset frequency, which is used as an original excitation signal.
[0076] Specifically, the signal generation module generates a 5-cycle sinusoidal signal with a sampling frequency fs of 200kHz and a center frequency fc of a preset frequency, which is used as the original excitation signal sig(n); wherein the 5-cycle sinusoidal signal is generated by using the Hanning window modulation method. After the original excitation signal sig(n) is phase-encoded using the Barker code, the coded excitation signal sig is obtained. e (n).
[0077] S12: Use Barker code to phase encode the original excitation signal to obtain a coded excitation signal.
[0078] Specifically, using Barker code to phase encode the excitation signal can compress a wider excitation signal into a narrow pulse signal with stronger amplitude after matched filtering, thereby increasing the average power of the excitation signal. e (n) carries more energy. Preferably, a 13-bit Barker code is used to phase encode the original excitation signal sig(n) to obtain the coded excitation signal sig e (n).
[0079] In some examples, the coded excitation signal is converted from an electrical signal into an acoustic wave and coupled to a rail so that the acoustic wave propagates along the rail, specifically including:
[0080] S21: power amplifying the coded excitation signal, converting the power-amplified coded excitation signal from an electrical signal into an acoustic wave, coupling it to the rail, so that the acoustic wave propagates along the rail.
[0081] Specifically, the signal coupling module is used to encode the excitation signal sig e (n) Perform power amplification to obtain the coded excitation signal sig after power amplification e (n) The electrical signal is converted into an acoustic wave, which is coupled to the rail and propagates along the rail. e (n) Power amplification can increase the energy of the sound waves coupled to the rails.
[0082] In some examples, receiving an acoustic wave at a first position, generating a first waveguide signal, preprocessing the first waveguide signal, and determining a track environment adaptive matched filter according to the preprocessed first waveguide signal specifically includes:
[0083] S31: receiving a sound wave at a first position, generating a first waveguide signal, and performing normalization processing on the first waveguide signal.
[0084] Specifically, the first sensor of the first signal processing module receives the sound wave at the first position and converts it into a first guided wave signal s 1_e (n), the first acquisition instrument acquires the first waveguide signal s generated by the first sensor 1_e (n), and the first waveguide signal s 1_e (n) is sent to the first operation unit, and the first waveguide signal s 1_e (n) is normalized. It is understandable that other signal processing methods can also be used to process the first waveguide signal s 1_e (n) is processed, which is only an example and is not specifically limited.
[0085] S32: Perform time domain flipping on the normalized first waveguide signal to determine the track environment adaptive matching filter.
[0086] Specifically, the first waveguide signal s 1_e (n) is received by the first sensor which is closer to the position where the acoustic wave is coupled by the signal coupling module, and the first waveguide signal s 1_e (n) includes part of the noise signal of the track environment, and the signal itself is not interfered by the noise signal of the track environment, so the first waveguide signal s 1_e (n) is processed to determine the orbit environment adaptive matching filter h e_mf (n).
[0087] In the embodiment of the present disclosure, among all linear filters, the matched filter outputs the largest peak signal-to-noise ratio, which only depends on the input signal energy and white noise power spectrum density, and has nothing to do with the input signal shape and noise distribution law. The matched filter can be used to identify the signal and compress the received signal into a single peak signal, so the matched filter is selected as the method for processing the waveguide signal in the present disclosure.
[0088] The matched filter includes a transfer function H(ω) and an impulse response h(t), and the input signal includes an input target signal s(t) and an input white noise n(t). The output y(t) of the filter can be expressed as: y(t) = s o (t)+n o (t);
[0089] Among them, the output target signal is s o (t), the output white noise is n o (t); output target signal s o The specific calculation formula of (t) is as follows:
[0090]
[0091] Where S(ω) is the frequency spectral density and ω is the angular frequency.
[0092] The power spectral density n0 of the input white noise s(t) multiplied by the square of the modulus of the transfer function H(ω) equals the output white noise s o The power spectral density of (t), that is, the output white noise s o The average power N0 of (t); the specific calculation formula is:
[0093]
[0094] At sampling time t0, the target signal s is output o The instantaneous power of (t) and the output white noise s o The ratio of the average power of (t) is r0; the specific calculation formula is as follows:
[0095]
[0096] According to the Cauchy-Schwarz inequality, r0 is calculated as follows:
[0097]
[0098] Where E is the energy of the target signal.
[0099] when When r0 can reach the maximum value, S * (ω) is the complex conjugate of S(ω). At this time, H(ω) is the optimal transfer function, which is equal to the complex conjugate of the target signal s(t). Where k is an arbitrary constant.
[0100] right By performing an inverse fast Fourier transform, the time domain impulse response h(t) of the matched filter can be obtained as:
[0101] h(t)=ks * (t0-t);
[0102] That is, the impulse response of the matched filter is the signal s(-t) determined after the target signal s(t) is time-domain flipped. Wherein, k is an arbitrary constant.
[0103] In the embodiment of the present disclosure, the first waveguide signal s 1_e (n) is the target signal in the above derivation, and the normalized first waveguide signal s 1_e '(n) Perform time domain flipping to determine the orbit environment adaptive matching filter h e_mf (n).
[0104] In some examples, receiving the sound wave at the second position, generating a second waveguide signal, performing matched filtering on the second waveguide signal using a track environment adaptive matched filter to obtain a rail detection signal, and detecting whether the rail is broken according to the rail detection signal, specifically includes:
[0105] S41: Receive the sound wave at the second position, generate a second waveguide signal, perform time domain convolution on the track environment adaptive matched filter and the second waveguide signal, and obtain a rail detection signal.
[0106] Specifically, the second sensor of the second signal processing module receives the sound wave at the second position and converts it into a second guided wave signal s 2_e (n), the second acquisition instrument collects the second waveguide signal s generated by the second sensor 2_e (n), and the second waveguide signal s 2_e (n) is sent to the second operation module, and the second waveguide signal s 2_e (n) and orbit environment adaptive matched filter h e_mf (n) Perform time domain convolution and the result is used as the rail detection signal y e_mf (n).
[0107] S42: According to a preset algorithm or a preset detection process, using the rail detection signal, detect whether the rail is broken.
[0108] Specifically, the rail detection signal y e_mf (n) In the second signal processing module, a preset algorithm is used to detect whether the rail is broken; or the rail detection signal y e_mf (n) Execute the preset detection process; for example, the rail detection signal y e_mf (n) performing signal filtering and determining the signal amplitude after signal filtering. When the signal amplitude is lower than a preset threshold, it is considered that the rail is broken. The second position may be a position at the other end of the range between two sections of the railway, opposite to the signal coupling module, that is, the propagation distance of the sound wave received by the second sensor is the distance between the two sections of the railway.
[0109] In some examples, the rail break detection method further includes: collecting a noise signal noise(n) of the track environment, and matching the noise signal noise(n) with the track environment adaptive matching filter h e_mf (n) Perform matched filtering to obtain the noise signal noise after track environment adaptive matched filtering e_mf (n), used to compare with the rail detection signal y e_mf(n) Calculate the signal-to-noise ratio (SNR). At the same time, the coding method of the excitation signal or the matched filter can be changed to determine the signal-to-noise ratio for comparison. Based on the comparison results of the calculated signal-to-noise ratio (SNR) and the signal-to-noise ratio determined by other methods, the signal-to-noise ratio gain effects of different coding methods or different Barker codes can be quantitatively analyzed, and different coding methods or different Barker codes can be selected in different orbital intervals according to the actual environment.
[0110] In the embodiment of the present disclosure, it is preferred to use a 13-bit Barker code to phase encode the original excitation signal sig(n) to obtain a coded excitation signal sig carrying more energy and higher average power. e (n). The coded excitation signal sig e (n) Convert the electrical signal into sound waves and couple them to the rails; receive the sound waves carrying the track environment noise and generate the first guided wave signal s 1_e (n), determine the orbit environment adaptive matching filter h according to the first waveguide signal e_mf (n). Receive the sound wave with a propagation distance equal to the distance between two sections of the railway, and generate a second waveguide signal s 2_e (n), the second waveguide signal s 2_e (n) The propagation time and distance on the track are longer, so it carries more environmental noise. At this time, the track environment adaptive matching filter h is used e_mf (n) for the second waveguide signal s 2_e (n) Perform matched filtering to improve the matched filtering effect and obtain a rail detection signal y e_mf (n). The signal-to-noise ratio of the signal is higher and the influence of the track environment is greatly eliminated. e_mf (n) and excludes the influence of the track environment on the rail detection signal y e_mf (n) influence, therefore, it can be used for long-distance rail break detection; the maximum effective detection distance in the prior art is about 2km, while the method in the present disclosure can achieve an effective detection distance of 5km to 10km.
[0111] In order to more clearly illustrate a rail break detection method in an embodiment of the present disclosure, a specific embodiment is provided below.
[0112] S101: The signal generation module uses a Hanning window modulation method to generate a 5-cycle sinusoidal signal with a sampling frequency of 200 kHz and a center frequency of a preset frequency, which is used as an original excitation signal sig(n).
[0113] S102: The signal generation module uses a 13-bit Barker code to phase encode the original excitation signal sig(n) to obtain a coded excitation signal sig e (n).
[0114] S103: The signal coupling module encodes the excitation signal sig e (n) Perform power amplification to obtain the coded excitation signal sig after power amplification e (n) The electrical signal is converted into sound waves, coupled to the rails, and the sound waves propagate along the rails.
[0115] S104: The first sensor of the first signal processing module receives the sound wave at the first position and generates a first guided wave signal s 1_e (n).
[0116] S105: The first computing unit of the first signal processing module processes the first waveguide signal s 1_e (n) Perform normalization processing, and convert the normalized first waveguide signal s 1_e '(n) Perform time domain flipping to determine the orbit environment adaptive matching filter h e_mf (n); the specific formula is as follows:
[0117] s 1_e '(n)=(s 1_e (n)-min(s 1_e (n))) / (max(s 1_e (n))-min(s 1_e (n)));
[0118] h e_mf (n) = s 1_e '(-n).
[0119] S106: The second sensor of the second signal processing module receives the sound wave at the second position and generates a second guided wave signal s 2_e (n).
[0120] S107: The second computing unit of the second signal processing module uses the track environment adaptive matching filter h e_mf (n) and the second waveguide signal s 2_e (n) Perform time domain convolution to obtain the rail detection signal y e_mf (n); the specific formula is as follows:
[0121] y e_mf (n) = s 2_e (n)*h e_mf (n).
[0122] S108: Using the rail detection signal y according to a preset algorithm or preset detection process e_mf (n), check whether the rails are broken.
[0123] In order to verify a rail break detection method disclosed in the present invention, a specific embodiment is provided below for verifying the above method.
[0124] like Figure 2 As shown in the figure, an experimental environment is built on a 500-meter-long free rail. The material of the free rail is the same as that of the actual rail. The experimental environment completely simulates the environment of the actual rail. The transducer of the signal coupling module is installed at one end of the rail to convert the coded excitation signal from an electrical signal to an acoustic signal and couple it to the rail. The sensor of the signal processing module is installed at the other end to receive the acoustic wave and generate a waveguide signal. The experimental environment also includes a collector for collecting the waveguide signal generated by the sensor, and a computer is used as a signal operation unit to process the waveguide signal accordingly.
[0125] like Figure 3 As shown, the signal generation module generates a 5-cycle sinusoidal signal modulated by a Hanning window with a sampling frequency fs of 200 kHz and a center frequency fc of a preset frequency, which is used as the original excitation signal sig(n).
[0126] like Figure 4 , 5 As shown, the original excitation signal sig(n) is phase-encoded using a 13-bit Barker code to obtain the coded excitation signal sig e (n). The coded excitation signal sig e (n) The electrical signal is converted into sound waves through the signal coupling module, and the transducer couples the sound waves to the rails. The sound waves propagate along the rails and are received by the sensor.
[0127] like Figure 6 , 7 As shown, Figure 6 It is the original excitation signal without coding, coupled to the signal waveform of the direct wave and the primary echo received by the sensor on the free track; Figure 7 It is the coded excitation signal after phase encoding using 13-bit Barker code, coupled to the signal waveform of the direct wave and the primary echo received by the sensor on the free track.
[0128] The sensor receives the direct wave of the sound wave coupled to the rail and converts it into the first guided wave signal s 1_e (n), the first waveguide signal s of the sensor is collected by the collector 1_e (n), the first waveguide signal s 1_e (n) sent to the computer; the transducer is coupled to the sound wave on the rail, which continues to propagate on the rail between the sensor and the transducer. When the sensor receives the direct wave, the sound wave propagates to the end of the rail, reflects the echo and propagates to the location of the transducer until the sound wave reaches the end of the rail again and reflects the echo and propagates to the location of the sensor. The sensor receives an echo returned from the location of the transducer and converts it into a second waveguide signal s 2_e (n), the second waveguide signal s 2_e(n) is sent to the computer. The first waveguide signal s 1_e (n) Perform normalization processing, and convert the normalized first waveguide signal s 1_e '(n), perform time domain flipping to obtain the orbit environment adaptive matched filter h e_mf (n); the specific formula is as follows:
[0129] s 1_e '(n)=(s 1_e (n)-min(s 1_e (n))) / (max(s 1_e (n))-min(s 1_e (n)));
[0130] h e_mf (n) = s 1_e '(-n).
[0131] like Figure 8 As shown, the second waveguide signal s 2_e (n) Adaptive matching filter h with orbit environment e_mf (n) Perform time domain convolution to obtain the rail detection signal y e_mf (n), the rail detection signal is the waveguide signal after track environment adaptive matching filtering; the second waveguide signal s 2_e (n) Adaptive matching filter h with orbit environment e_mf (n) The specific formula for time domain convolution is as follows:
[0132] y e_mf (n) = s 2_e (n)*h e_mf (n).
[0133] like Fig. 9 As shown, the noise signal noise(n) near the track is collected, and the noise signal noise(n) is matched with the track environment adaptive matching filter h e_mf (n) Perform time domain convolution to obtain the noise signal noise after track environment adaptive matched filtering e_mf (n); the specific formula is as follows:
[0134] noise e_mf (n) = noise(n)*h e_mf (n).
[0135] Noise signal noise after adaptive matching filtering according to the track environment e_mf (n) and rail detection signal y e_mf (n), determine the signal-to-noise ratio SNR; the specific formula is as follows:
[0136] SNR=20*log 10(max(y e_mf (n)) / max(noise e_mf (n))).
[0137] The uncoded original excitation signal sig(n) is converted from an electrical signal to an acoustic wave through a signal coupling module, and the transducer couples the acoustic wave to the rail, and the acoustic wave propagates along the rail. The sensor receives the direct wave of the acoustic wave coupled to the rail and converts it into a second comparison signal s1(n). The second comparison signal s1(n) of the sensor is collected by the acquisition instrument, and the first comparison signal s1(n) is sent to the computer; the acoustic wave coupled to the rail by the transducer continues to propagate on the rail between the sensor and the transducer. When the sensor receives the direct wave, the acoustic wave propagates to the end of the rail, and the echo is reflected and propagated to the location of the transducer, until the acoustic wave reaches the end of the rail again and the echo is reflected and propagated to the location of the sensor. The sensor receives an echo returned from the location of the transducer, converts it into a second comparison signal s2(n), and sends the second comparison signal s2(n) to the computer.
[0138] The noise signal noise(n) near the track is collected, and the signal-to-noise ratio SNR1 is calculated according to the second comparison signal s2(n) and the noise signal noise(n); the specific formula is as follows:
[0139] SNR1=20*log 10(max(s2(n)) / max(noise(n))).
[0140] like Fig.10 As shown, in order to combine the traditional matched filter with the track environment adaptive matched filter h e_mf (n) for comparison, the coded excitation signal sig e (n) Directly perform time domain flipping to obtain the traditional matched filter h mf (n); the specific formula is as follows:
[0141] h mf (n) = sig e (-n).
[0142] The second waveguide signal s 2_e (n) and the traditional matched filter h mf (n) Perform time domain convolution to obtain the second waveguide signal y after traditional matched filtering mf (n); the specific formula is as follows:
[0143] y mf (n) = s 2_e (n)*h mf (n).
[0144] The collected noise signal noise(n) is also filtered using the traditional matched filter h mf (n) Perform matched filtering to obtain the noise signal noise after traditional matched filtering mf (n); the specific formula is as follows:
[0145] noise mf (n) = noise(n)*h mf (n).
[0146] According to the noise signal noise after traditional matched filtering mf (n) and the second waveguide signal y after traditional matched filtering mf (n), determine the signal-to-noise ratio SNR2; the specific formula is as follows:
[0147] SNR2=20*log10(max(y mf (n)) / max(noise mf (n))).
[0148] like Fig.11 As shown, from left to right, the signal-to-noise ratio SNR1 of the second comparison signal s2(n) and the noise signal noise(n), the noise signal noise after the traditional matched filtering mf (n) and the second waveguide signal y after traditional matched filtering mf (n) SNR2, noise signal after environmental adaptive matching filtering e_mf (n) and rail detection signal y e_mf As can be seen from the figure, the second comparison signal s2(n) is the waveguide signal corresponding to the uncoded original excitation signal sig(n), and the signal-to-noise ratio SNR1 of the second comparison signal s2(n) and the noise signal noise(n) is the smallest; accordingly, the second waveguide signal s 2_e (n) is the coded excitation signal sig for phase encoding of 13-bit Barker code e (n) The corresponding waveguide signal, the noise signal after traditional matched filtering mf (n) and the second waveguide signal y after traditional matched filtering mf The signal-to-noise ratio SNR2 of (n) is significantly improved compared with SNR1; further, the rail detection signal y e_mf (n) is the second waveguide signal s 2_e (n) Using orbital environment adaptive matched filter h e_mf (n) The noise signal obtained by matching filtering, the noise after the track environment adaptive matching filtering e_mf (n) and rail detection signal y e_mfThe signal-to-noise ratio (SNR) of (n) is improved compared with SNR2, and the SNR is improved by about 24dB compared with SNR1, which can effectively increase the distance of rail break detection in the area.
[0149] In a second aspect, the disclosed embodiment further provides a rail break detection system, the system comprising: a signal generating module, a signal coupling module, a first signal processing module and a second signal processing module; the signal generating module is used to generate an original excitation signal, and phase encode the original excitation signal to obtain a coded excitation signal; the signal coupling module is used to convert the coded excitation signal from an electrical signal into an acoustic wave, and couple it to the rail so that the acoustic wave propagates along the rail; the first signal processing module is used to receive the acoustic wave at a first position, generate a first waveguide signal, preprocess the first waveguide signal, and determine a track environment adaptive matching filter according to the preprocessed first waveguide signal; the second signal processing module is used to receive the acoustic wave at a second position, generate a second waveguide signal, match filter the second waveguide signal using a track environment adaptive matching filter, and obtain a rail detection signal, and detect whether the rail is broken according to the rail detection signal.
[0150] In some examples, the signal coupling module includes: a transducer and a power amplifier; the power amplifier is used to amplify the power of the coded excitation signal; the transducer is used to convert the power-amplified coded excitation signal from an electrical signal to an acoustic wave and couple it to the rail.
[0151] In some examples, the first signal processing module includes a first sensor, a first collector and a first operation unit; the first sensor is used to receive sound waves at a first position and generate a first waveguide signal; the first collector is used to collect the first waveguide signal of the first sensor and send it to the first operation unit; the first operation unit is used to normalize the first waveguide signal, flip the normalized first waveguide signal in the time domain, and determine the orbital environment adaptive matching filter.
[0152] In some examples, the second signal processing module includes a second sensor, a second collector and a second operation unit; the second sensor is used to receive sound waves at a second position and generate a second waveguide signal; the second collector is used to collect the second waveguide signal of the second sensor and send it to the second operation unit; the second operation unit is used to perform time domain convolution on the track environment adaptive matching filter and the second waveguide signal to obtain a rail detection signal, and use the rail detection signal to detect whether the rail is broken according to a preset algorithm or a preset detection process.
[0153] It should be understood that the first computing unit and the second computing unit can be the same physical device or system, or different physical devices or systems. The specific setting method is adjusted according to the actual distance between the two sections of the railway.
[0154] Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure.
Claims
1. A rail break detection method, characterized in that: The method comprises: Generate an original excitation signal, and perform phase encoding on the original excitation signal to obtain an encoded excitation signal; Converting the coded excitation signal from an electrical signal into an acoustic wave, and coupling it to the rail, so that the acoustic wave propagates along the rail; receiving an acoustic wave at a first position, generating a first waveguide signal, preprocessing the first waveguide signal, and determining a track environment adaptive matched filter according to the preprocessed first waveguide signal; Receive the sound wave at the second position, generate a second waveguide signal, use the track environment adaptive matched filter to perform matched filtering on the second waveguide signal to obtain a rail detection signal, and detect whether the rail is broken based on the rail detection signal.
2. The method according to claim 1, characterized in that The generating of the original excitation signal and performing phase encoding on the original excitation signal to obtain the encoded excitation signal comprises: Generate a 5-period sinusoidal signal with a sampling frequency of 200 kHz and a center frequency of a preset frequency, which is used as the original excitation signal; The original excitation signal is phase-encoded using a Barker code to obtain the coded excitation signal.
3. The method according to claim 1, characterized in that The step of converting the coded excitation signal from an electrical signal into an acoustic wave and coupling it to the rail so that the acoustic wave propagates along the rail comprises: The coded excitation signal is power-amplified, and the power-amplified coded excitation signal is converted from an electrical signal to an acoustic wave, which is coupled to a steel rail so that the acoustic wave propagates along the steel rail.
4. The method according to claim 1, characterized in that: The method of receiving the sound wave at the first position, generating a first waveguide signal, preprocessing the first waveguide signal, and determining a track environment adaptive matching filter according to the preprocessed first waveguide signal includes: receiving the sound wave at the first position, generating the first waveguide signal, and performing normalization processing on the first waveguide signal; The first waveguide signal after normalization is flipped in time domain to determine the track environment adaptive matching filter.
5. The method according to claim 1, characterized in that The receiving of the sound wave at the second position, generating a second waveguide signal, using a track environment adaptive matched filter to perform matched filtering on the second waveguide signal to obtain a rail detection signal, and detecting whether the rail is broken according to the rail detection signal, comprises: Receiving the sound wave at the second position, generating the second waveguide signal, performing time domain convolution on the track environment adaptive matched filter and the second waveguide signal, and obtaining the rail detection signal; According to a preset algorithm or a preset detection process, the rail detection signal is used to detect whether the rail is broken.
6. The method according to claim 2, characterized in that The generating of a 5-cycle sinusoidal signal with a sampling frequency of 200 kHz and a center frequency of a preset frequency, which is used as an original excitation signal, includes: Use the Hanning window modulation method to generate a 5-cycle sine signal.
7. The method according to claim 2, characterized in that The method of using Barker code to phase encode the original excitation signal to obtain the coded excitation signal includes: The original excitation signal is phase-encoded using a 13-bit Barker code to obtain the coded excitation signal.
8. The method according to claim 1, characterized in that The method further comprises: The noise signal of the track environment is collected and used to determine the signal-to-noise ratio with the rail detection signal.
9. A rail break detection system, characterized in that: The system comprises: a signal generating module, a signal coupling module, a first signal processing module and a second signal processing module; The signal generating module is used to generate an original excitation signal, and perform phase encoding on the original excitation signal to obtain an encoded excitation signal; The signal coupling module is used to convert the coded excitation signal from an electrical signal into an acoustic wave, and couple it to the rail so that the acoustic wave propagates along the rail; The first signal processing module is used to receive the sound wave at the first position, generate a first waveguide signal, preprocess the first waveguide signal, and determine the track environment adaptive matched filter according to the preprocessed first waveguide signal; The second signal processing module is used to receive the sound wave at the second position, generate a second waveguide signal, use the track environment adaptive matched filter to perform matched filtering on the second waveguide signal to obtain a rail detection signal, and detect whether the rail is broken based on the rail detection signal.
10. The system according to claim 9, characterized in that The signal coupling module includes: a transducer and a power amplifier; The power amplifier is used to amplify the power of the coded excitation signal; The transducer is used to convert the power-amplified coded excitation signal from an electrical signal into an acoustic wave, and couple it to the rail.
11. The system according to claim 9, characterized in that The first signal processing module includes: a first sensor, a first acquisition instrument and a first computing unit; The first sensor is used to receive the sound wave at the first position and generate the first guided wave signal; The first acquisition instrument is used to collect the first waveguide signal of the first sensor and send it to the first computing unit; The first operation unit is used to normalize the first waveguide signal, perform time domain flipping on the normalized first waveguide signal, and determine the track environment adaptive matching filter.
12. The system according to claim 9, characterized in that The second signal processing module includes a second sensor, a second acquisition instrument and a second computing unit; The second sensor is used to receive the sound wave at the second position and generate the second guided wave signal; The second collector is used to collect the second waveguide signal of the second sensor and send it to the second computing unit; The second operation unit is used to perform time domain convolution on the track environment adaptive matched filter and the second waveguide signal to obtain the rail detection signal, and use the rail detection signal to detect whether the rail is broken according to a preset algorithm or a preset detection process.
Citation Information
Cited By
Steel rail defect identification method and device, storage medium and electronic equipment
CN120195278A
