A multi-scene adaptive distributed optical fiber vibration intelligent monitoring method and system

Abstract

The application relates to a distributed optical fiber vibration intelligent monitoring method and system suitable for multiple scenes, which comprises the following steps: detecting light intensity data along a sensing optical cable; demodulating the light intensity data in an effective length to obtain vibration signal data at the sensing optical cable; then analyzing time domain peak energy characteristic values, signal envelope energy characteristic values and main frequency band energy characteristic values of the vibration signal data; then adaptively and dynamically configuring time domain peak energy characteristic threshold values, signal envelope energy characteristic threshold values and main frequency band energy characteristic threshold values of each sensing unit; jointly using the time domain peak energy characteristic values, the signal envelope energy characteristic values and the main frequency band energy characteristic values to distinguish external disturbance events and determine whether the events are disturbance event signals; and finally performing pattern recognition on vibration signal types corresponding to the disturbance events. The application uses the complementarity of different energy characteristic values, comprehensively analyzes target signals in multiple characteristic quantities and multiple dimensions, and improves the accuracy and robustness of vibration recognition.

Description

Technical Field

[0001] This application relates to the field of optical fiber sensing technology, and in particular to a distributed optical fiber vibration intelligent monitoring method and system adaptable to multiple scenarios. Background Technology

[0002] With the development of optical communication technology, optical sensing technology has been greatly advanced. Distributed fiber optic sensing technology can continuously sense and locate physical quantities along the fiber optic cable in real time, and it has attracted increasing attention from researchers, resulting in many technological breakthroughs. Due to its high sensitivity, large spatial dynamic range, and high positioning accuracy, it has become the mainstream technology in the field of distributed fiber optic sensing.

[0003] In practical monitoring applications, due to the large range of communication optical cables and the alternation of various complex scenarios, such as soil, cement, sand, bridges, tunnels, roads, and fences, how to achieve seamless connection in various complex scenarios based on distributed optical fiber vibration sensing systems and provide comprehensive and integrated monitoring services has become a challenging problem.

[0004] In related technologies, the sensing performance of distributed fiber optic vibration sensing systems in different scenarios is affected by differences in the coupling installation between the optical cable and the external structure and changes in the monitoring environment. This effect manifests as the same type of vibration event signal exhibiting inconsistent distribution characteristics in different scenarios.

[0005] If the traditional method of setting a unified threshold for the entire link is adopted, although the parameter threshold setting operation is simple, the environmental conditions, signal characteristics and monitoring requirements of different scenarios are significantly different. The unified threshold setting method often cannot meet the monitoring needs of multiple scenarios.

[0006] While the segmented threshold setting method can meet the needs of disturbance event identification in different scenarios by setting thresholds in segments, it requires separate configuration for each scenario or signal segment, which is cumbersome and consumes a lot of time and manpower.

[0007] Currently, most distributed fiber optic vibration sensing and monitoring technologies mainly use a single feature domain energy threshold to identify external disturbance events. However, relying on a single feature domain energy threshold can easily lead to misjudgment of real disturbance events in complex monitoring scenarios, affecting the accuracy of the monitoring system. Summary of the Invention

[0008] This application provides a distributed optical fiber vibration intelligent monitoring method and system adapted to multiple scenarios, in order to solve the problem in related technologies that the energy threshold of a single feature domain is prone to misjudging real disturbance events in complex monitoring scenarios, thus affecting the accuracy of the monitoring system.

[0009] The first aspect of this application provides a distributed optical fiber vibration intelligent monitoring method adaptable to multiple scenarios, the method comprising the following steps: The light intensity data of the backscattered Rayleigh light signal in the optical fiber caused by the vibration source along the detection sensing optical cable; The effective length of the sensing optical cable is determined based on the light intensity data. The light intensity data within the effective length is demodulated to obtain the vibration signal data along the sensing optical cable. The energy characteristic values ​​of the vibration signal data are analyzed, including the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit. The time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold of each sensing unit are adaptively and flexibly configured dynamically. The external disturbance event is identified by combining the time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics. The vibration signal type identified as an external disturbance event will be used for pattern recognition.

[0010] In some embodiments: determining the effective length of the sensing optical cable based on light intensity data specifically includes: Set light intensity threshold I th , The backscattered Rayleigh light signals detected at each monitoring point within the time interval t are: ; The light intensity information of each monitoring point was calculated according to Formula 1. X 1, X 2,…, X n ]; Formula 1 in: x ij Representing the i Time of the first j The light intensity values ​​corresponding to each monitoring point; Compare the light intensity values ​​at each monitoring point with the light intensity threshold. I th By comparison, the first signal in this set of data that exceeds the light intensity threshold is identified. I th The data value is denoted as X p And index the last one that is greater than the light intensity threshold. I th The data value is denoted as X qThe starting point of the sensing optical cable can be calculated according to Formula 2. L 1 and termination point L 2:

[0011] in: Fs The sampling rate of the data acquisition card. n f The refractive index coefficient of the sensing optical cable. c At the speed of light, p The first one greater than the light intensity threshold I th Number of data points q The last one is greater than the light intensity threshold. I th Number of data points; The effective length of the sensing optical cable is Intensity demodulation technology is used to demodulate the intensity of each data point within the effective length of the sensing optical cable, and finally the vibration signal data corresponding to each sensing unit is obtained.

[0012] In some embodiments, analyzing vibration signal data to obtain time-domain peak energy characteristic values ​​includes: Perform time-domain segmentation on the target signal, dividing the original time-domain signal into... m Each time domain segment is configured with a corresponding time domain peak-finding width. d 1, d 2,…, d m and peak-seeking threshold [ h 1, h 2,…, h m ]; Peak finding is performed on the positive and negative domains of the data amplitude for each time domain segment, where the first peak is... i The set of peaks found in the positive value region of each time domain segment is denoted as... Next, take the negative value of the target signal and perform peak finding in its negative range. The set of peaks found in the negative range is denoted as... ; Calculate the first [formula] using Formula 3. i The median of the peak set found for each time domain segment is denoted as the peak energy corresponding to that time domain segment. ; Formula 3 Formula 4 Finally, this can be calculated using Formula 4. m The average of the peak energies of each time-domain segment is denoted as the time-domain peak energy characteristic value corresponding to the target signal. P .

[0013] In some embodiments, analyzing vibration signal data to obtain signal envelope energy characteristic values ​​includes: Set the width of the time-domain sliding window. d The sliding window moves one step at a time. l Assuming the target signal data is [ y 1, y 2,…, y N According to Formula 5, the energy value of each sliding window is calculated by energy integration, and the integral energy curve is obtained and denoted as follows. ;

[0014] Linear spline interpolation is used to perform interval linear interpolation on the data points in the integral energy curve. The number of data points interpolated is... The number of data points in the interpolated integral energy curve is the same as the number of data points in the original target signal, denoted as... ; Based on the calculated time-domain peak energy characteristic value P Set the threshold for the integral energy curve. P * d ,in: d As the weighting coefficient, sequentially with the threshold of the integral energy curve P * d Compare and find the data segment with the longest continuous value greater than the threshold of the integral energy curve, denoted as . ; Formula Six Finally, the signal envelope energy characteristic value S corresponding to the target signal is calculated according to Formula 6.

[0015] In some embodiments, analyzing vibration signal data to obtain the main frequency band energy characteristic value includes: The spectral information corresponding to the target signal data is obtained through Fourier transform, and the maximum value of the signal spectral energy is found. Y max ; Set the main frequency band energy characteristic discrimination parameters a * Y max The signal spectrum data is sequentially compared with the main frequency band energy characteristic discrimination parameters. a * Y max Compare; Find all signals in the spectrum corresponding to this signal that are greater than [a certain value]. a * Y max The frequency value point corresponding to the frequency domain energy is denoted as . ; Find the maximum frequency point according to Formula 7. With minimum frequency point The difference is taken as the dominant frequency band of the signal: Formula 7 Formula 8 Then, by integrating all spectral energy values ​​in the main frequency band of the signal, the energy characteristic value of the main frequency band corresponding to the signal is obtained according to Formula 8. F .

[0016] In some embodiments: the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold of each sensing unit are adaptively and flexibly configured dynamically, specifically including: Based on the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit, the corresponding time-domain peak energy characteristic curve of each sensing unit is obtained. G P Signal envelope energy characteristic curve G S and the main frequency band energy characteristic curve G F ; Polynomial curve fitting is performed on each energy characteristic curve using Formula 9 to achieve adaptive setting of the domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold parameters corresponding to each sensing unit.

[0017] Set the corresponding adjustment curvature scaling factor and curve bias factor for the fitted curve respectively;

[0018] Based on the formula, fit the time-domain peak energy characteristic curve. Signal envelope energy characteristic fitting curve Fitting curve of energy characteristics of the main frequency band Make dynamic adjustments.

[0019] In some embodiments, it also includes: Based on real-time data and changes in the current scene environment, flexible single-point threshold corrections can be performed on the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for certain abnormal sensing unit points or junctions between different scenes, resulting in optimized time-domain peak energy characteristic thresholds for each monitoring point. Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold ; Using the time-domain peak energy characteristic threshold of each sensing unit Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Disturbance detection is performed on the external signals detected at each monitoring point.

[0020] In some embodiments, the combined use of time-domain peak energy characteristics, signal envelope energy characteristics, and dominant frequency band energy characteristics to determine external disturbance events specifically includes: Thresholds of peak energy characteristics in the time domain Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Set the corresponding weight parameters The disturbance discrimination formula eleven is as follows: The disturbance event discrimination formula is based on the comprehensive information of time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics.

[0021] in, The first i The time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of the signal detected at each monitoring point; If the time-domain peak energy characteristic value of the current target signal P Signal envelope energy eigenvalues S and the energy characteristic value of the main frequency band F At the same time greater than If the condition is met, the signal is identified as an abnormal disturbance event; otherwise, the signal is identified as a non-abnormal disturbance event.

[0022] In some embodiments: pattern recognition is performed on the vibration signal type identified as corresponding to an external disturbance event, specifically including: The event type of the disturbance event is identified by using the original signal waveform data corresponding to the identified disturbance event and its corresponding time-domain peak energy feature value, signal envelope energy feature value and main frequency band energy feature value. A neural network model is built to perform convolution calculations on the original signal waveform data corresponding to the disturbance event and extract the corresponding waveform features; The waveform features are combined with the extracted time-domain peak energy features, signal envelope energy features, and main frequency band energy features to form a multi-dimensional feature matrix. Finally, the matrix is ​​fed into a pattern classifier for matrix calculation to classify and identify the types of disturbance events.

[0023] The second aspect of this application provides a distributed fiber optic vibration intelligent monitoring system adaptable to multiple scenarios, including: A distributed optical fiber vibration detection module is used to detect the light intensity data of the backscattered Rayleigh light signal in the optical fiber caused by the vibration source along the sensing optical cable. The optical fiber sensing cable length calibration module is used to calibrate the effective length of the optical fiber sensing cable based on light intensity data, demodulate the light intensity data within the effective length, and obtain vibration signal data along the optical fiber sensing cable. The signal energy characteristic analysis module is used to analyze the energy characteristic values ​​of vibration signal data. The energy characteristic values ​​include the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit. An adaptive dynamic adjustment module for each energy characteristic threshold parameter is used to adaptively and flexibly configure the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for each sensing unit. Based on the multi-feature energy threshold disturbance discrimination module, it is used to judge external disturbance events by combining the time-domain peak energy features, signal envelope energy features and main frequency band energy features of each sensing unit, and to determine whether it is an external disturbance event signal. The vibration event pattern recognition module is used to perform pattern recognition on the vibration signal type corresponding to the external disturbance event.

[0024] The beneficial effects of the technical solution provided in this application include: This application provides a distributed optical fiber vibration intelligent monitoring method and system adapted to multiple scenarios. The distributed optical fiber vibration intelligent monitoring method of this application first detects the light intensity data of the backscattered Rayleigh light signal in the optical fiber caused by the vibration source along the sensing optical cable; secondly, it calibrates the effective length of the sensing optical cable based on the light intensity data, demodulates the light intensity data within the effective length, and obtains the vibration signal data along the sensing optical cable. Next, the energy characteristic values ​​of the vibration signal data are analyzed. These energy characteristic values ​​include the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit. Then, the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold of each sensing unit are adaptively and flexibly configured dynamically. Finally, the external disturbance event is judged by the combined time-domain peak energy characteristic, signal envelope energy characteristic, and main frequency band energy characteristic of each sensing unit to determine whether it is an external disturbance event signal. Finally, the vibration signal type corresponding to the external disturbance event is subjected to pattern recognition.

[0025] Therefore, the distributed optical fiber vibration intelligent monitoring method of this application obtains vibration signal data along the sensing optical cable by demodulating the light intensity data of the backscattered Rayleigh light signal caused by the vibration source within the optical fiber. By analyzing the energy characteristic values ​​of the vibration signal data, the energy characteristic values ​​are divided into time-domain peak energy characteristic values, signal envelope energy characteristic values, and dominant frequency band energy characteristic values. These three energy characteristic values ​​are compared and identified, and the complementarity of different energy characteristic values ​​is utilized to comprehensively analyze the target signal from multiple dimensions and features, thereby improving the accuracy and robustness of vibration identification. The adaptive dynamic adjustment technology of the energy threshold parameter can automatically optimize the threshold parameter configuration according to real-time data and environmental changes, flexibly responding to the needs of different scenarios and exhibiting strong adaptability and flexibility. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A structural block diagram of a distributed fiber optic vibration intelligent monitoring system adapted to multiple scenarios for this application; Figure 2 This is a time-domain peak energy characteristic distribution diagram corresponding to each sensing unit of this application; Figure 3 This is a distribution diagram of the signal envelope energy characteristics corresponding to each sensing unit in this application; Figure 4 This is a distribution diagram of the main frequency band energy characteristics corresponding to each sensing unit in this application; Figure 5 This is an adaptive dynamic setting diagram of the energy characteristic thresholds corresponding to each sensing unit provided in this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] This application provides a distributed optical fiber vibration intelligent monitoring method and system adapted to multiple scenarios, which can solve the problem in related technologies that the energy threshold of a single feature domain is prone to misjudging real disturbance events in complex monitoring scenarios, thus affecting the accuracy of the monitoring system.

[0030] The first aspect of this application provides a distributed optical fiber vibration intelligent monitoring method adaptable to multiple scenarios, the method comprising the following steps: Step 101: Detect the light intensity data of the backscattered Rayleigh light signal in the optical fiber caused by the vibration source along the sensing optical cable; Based on the Rayleigh scattering effect of optical fiber, when a laser pulse propagates in the optical fiber, external vibration can change the refractive index or strain of the optical fiber, causing changes in the phase or intensity of the scattered light. By detecting the light intensity data of the backscattered Rayleigh light signal, the location, frequency and amplitude of the vibration event can be calculated.

[0031] Step 102: Determine the effective length of the sensing optical cable based on the light intensity data. Demodulate the light intensity data within the effective length to obtain vibration signal data along the sensing optical cable. Using the sensing optical cable as a transmission medium, the surrounding environment can be sensed, enabling continuous monitoring of vibration signals over a range of hundreds of kilometers on a single optical fiber. The longer the sensing optical cable, the weaker the light signal energy received by the optical module connected to it. The effective length of the sensing optical cable is defined as the length of the position where the optical module can reach a set threshold of light intensity data.

[0032] Step 103: Analyze the energy characteristic values ​​of the vibration signal data. The energy characteristic values ​​include the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit. Since the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value are complementary, the comprehensive analysis of the target signal with multiple features and dimensions can significantly improve the accuracy and robustness of identification.

[0033] Step 104: Adaptively and dynamically configure the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for each sensing unit. Through adaptive dynamic adjustment technology of energy threshold parameters, the characteristic energy parameter thresholds corresponding to different sensing units are adaptively set. At the same time, flexible adjustments can be made for abnormal sensing unit points or junctions between different scenarios to further optimize the system threshold parameter configuration and improve the system's practicality and robustness in complex scenarios.

[0034] Step 105: Combine the time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics to identify external disturbance events and determine whether they are external disturbance event signals; compare the external signals detected by each sensing unit with the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold corresponding to each sensing unit to perform disturbance discrimination and determine whether the external signals detected by each sensing unit are external abnormal disturbance event signals.

[0035] Step 106: Perform pattern recognition on the vibration signal type corresponding to the external disturbance event. Calculate the original signal waveform data of the external vibration signal and its corresponding time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value. The type of the external disturbance event signal can be calculated, which is the event type that generates the external disturbance event signal.

[0036] In some alternative embodiments: This application provides a distributed optical fiber vibration intelligent monitoring method adapted to multiple scenarios, wherein the effective length of the sensing optical cable is determined based on light intensity data, specifically including: Set light intensity threshold I th ; The backscattered Rayleigh light signals detected at each monitoring point within the time interval t are: ; The light intensity information of each monitoring point was calculated according to Formula 1. X 1, X 2,…, X n ]; Formula 1 in: x ij Representing the i Time of the first j The light intensity values ​​corresponding to each monitoring point; Compare the light intensity values ​​at each monitoring point with the light intensity threshold. I th By comparison, the first signal in this set of data that exceeds the light intensity threshold is identified. I th The data value is denoted as X p And index the last one that is greater than the light intensity threshold. I th The data value is denoted as X q The starting point of the sensing optical cable can be calculated according to Formula 2. L 1 and termination point L 2:

[0037] in: Fs The sampling rate of the data acquisition card. n f The refractive index coefficient of the sensing optical cable. c At the speed of light, p The first one greater than the light intensity threshold I th Number of data points q The last one is greater than the light intensity threshold. I th Number of data points; The effective length of the sensing optical cable is Intensity demodulation technology is used to demodulate the intensity of each data point within the effective length of the sensing optical cable, and finally the vibration signal data corresponding to each sensing unit is obtained.

[0038] In some optional embodiments: This application provides a distributed optical fiber vibration intelligent monitoring method adapted to multiple scenarios, wherein the method analyzes vibration signal data to obtain time-domain peak energy characteristic values, including: Perform time-domain segmentation on the target signal, dividing the original time-domain signal into... m Each time domain segment is configured with a corresponding time domain peak-finding width. d 1, d 2,…, d m and peak-seeking threshold [ h 1, h 2,…, h m ]; Peak finding is performed on the positive and negative domains of the data amplitude for each time domain segment, where the first peak is... i The set of peaks found in the positive value region of each time domain segment is denoted as... Next, take the negative value of the target signal and perform peak finding in its negative range. The set of peaks found in the negative range is denoted as... ; Calculate the first [formula] using Formula 3. i The median of the peak set found for each time domain segment is denoted as the peak energy corresponding to that time domain segment. ; Formula 3 Formula 4 Finally, this can be calculated using Formula 4. m The average of the peak energies of each time-domain segment is denoted as the time-domain peak energy characteristic value corresponding to the target signal. P .

[0039] In some optional embodiments: This application provides a distributed optical fiber vibration intelligent monitoring method adapted to multiple scenarios, wherein the method analyzes vibration signal data to obtain signal envelope energy characteristic values ​​including: Set the width of the time-domain sliding window. d The sliding window moves one step at a time. l Assuming the target signal data is [ y 1, y 2,…, y N According to Formula 5, the energy value of each sliding window is calculated by energy integration, and the integral energy curve is obtained and denoted as follows. ;

[0040] Linear spline interpolation is used to perform interval linear interpolation on the data points in the integral energy curve. The number of data points interpolated is... The number of data points in the interpolated integral energy curve is the same as the number of data points in the original target signal, denoted as... ; Based on the calculated time-domain peak energy characteristic value P Set the threshold for the integral energy curve. P * d ,in: d As the weighting coefficient, sequentially with the threshold of the integral energy curve P * d Compare and find the data segment with the longest continuous value greater than the threshold of the integral energy curve, denoted as . ; Formula Six Finally, the signal envelope energy characteristic value S corresponding to the target signal is calculated according to Formula 6.

[0041] In some optional embodiments: This application provides a distributed optical fiber vibration intelligent monitoring method adapted to multiple scenarios, wherein the method analyzes vibration signal data to obtain the main frequency band energy characteristic value, including: The spectral information corresponding to the target signal data is obtained through Fourier transform, and the maximum value of the signal spectral energy is found. Y max ; Set the main frequency band energy characteristic discrimination parameters a * Y max The signal spectrum data is sequentially compared with the main frequency band energy characteristic discrimination parameters. a * Y max Compare; Find all signals in the spectrum corresponding to this signal that are greater than [a certain value]. a* Y max The frequency value point corresponding to the frequency domain energy is denoted as . ; Find the maximum frequency point according to Formula 7. With minimum frequency point The difference is taken as the dominant frequency band of the signal: Formula 7 Formula 8 Then, by integrating all spectral energy values ​​in the main frequency band of the signal, the energy characteristic value of the main frequency band corresponding to the signal is obtained according to Formula 8. F .

[0042] In some alternative embodiments: see Figures 2 to 5 As shown in the figure, this application provides a distributed optical fiber vibration intelligent monitoring method adaptable to multiple scenarios. The method adaptively and flexibly dynamically configures the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold of each sensing unit, specifically including: Based on the above embodiments, the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit are calculated, and the corresponding time-domain peak energy characteristic curve of each sensing unit is obtained. G P Signal envelope energy characteristic curve G S and the main frequency band energy characteristic curve G F ; Polynomial curve fitting is performed on each energy characteristic curve using Formula 9 to achieve adaptive setting of the domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold parameters corresponding to each sensing unit.

[0043] In order to obtain more accurate information on the threshold parameters of each energy feature, corresponding adjustment curvature scaling factor and curve bias factor are set for the fitted curves respectively.

[0044] Based on the formula, the time-domain peak energy characteristic fitting curves are obtained separately. Signal envelope energy characteristic fitting curve Fitting curve of energy characteristics of the main frequency band Make dynamic adjustments.

[0045] To better apply the energy threshold parameters to special points such as abnormal sensing unit points or junctions between different scenarios, single-point threshold corrections can be performed on the domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for each sensing unit at certain abnormal sensing unit points or junctions between different scenarios, based on real-time data and changes in the current scene environment. This yields optimized time-domain peak energy characteristic thresholds for each monitoring point. Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Further optimization of the configuration of threshold parameters for each energy characteristic is needed to improve the system's practicality and robustness in complex scenarios.

[0046] Using the time-domain peak energy characteristic threshold of each sensing unit Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Disturbance detection is performed on the external signals detected at each monitoring point.

[0047] In some alternative embodiments: see Figures 2 to 5 As shown in the figure, this application provides a distributed optical fiber vibration intelligent monitoring method adapted to multiple scenarios. The method uses time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics to determine external disturbance events. Specifically, it includes: To further improve the accuracy of identifying external disturbance events in complex scenarios, the peak energy feature threshold in the time domain was adjusted. Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Set the corresponding weight parameters The disturbance discrimination formula eleven is as follows: The disturbance event discrimination formula is based on the comprehensive information of time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics.

[0048] in, The first i The time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of the signal detected at each monitoring point; If the time-domain peak energy characteristic value of the current target signal P Signal envelope energy eigenvalues S and the energy characteristic value of the main frequency band F At the same time greater than If the condition is met, the signal is identified as an abnormal disturbance event; otherwise, the signal is identified as a non-abnormal disturbance event.

[0049] In some optional embodiments: This application provides a distributed optical fiber vibration intelligent monitoring method adapted to multiple scenarios. The method performs pattern recognition on the vibration signal type corresponding to the external abnormal disturbance event, specifically including: The event type of the disturbance event is identified by using the original signal waveform data corresponding to the disturbance event and its corresponding time-domain peak energy feature value, signal envelope energy feature value and main frequency band energy feature value.

[0050] A neural network model is built to perform convolution calculations on the original signal waveform data corresponding to the disturbance event and extract the corresponding waveform features. The waveform features are combined with the extracted time-domain peak energy feature values, signal envelope energy feature values ​​and main frequency band energy feature values ​​to form a multi-dimensional feature matrix. Finally, the matrix is ​​fed into a pattern classifier for matrix calculation to classify and identify the type of disturbance event.

[0051] See Figure 1 As shown, a second aspect of this application provides a distributed optical fiber vibration intelligent monitoring system adaptable to multiple scenarios, comprising: The distributed optical fiber vibration detection module is used to detect the light intensity data of the backscattered Rayleigh light signal in the optical fiber caused by the vibration source along the sensing optical cable. Based on the Rayleigh scattering effect of optical fiber, when a laser pulse propagates in the optical fiber, external vibration can change the refractive index or strain of the optical fiber, causing changes in the phase or intensity of the scattered light. By detecting the light intensity data of the backscattered Rayleigh light signal, the location, frequency and amplitude of the vibration event can be calculated.

[0052] The optical fiber sensing cable length calibration module is used to calibrate the effective length of the optical fiber sensing cable based on light intensity data. It demodulates the light intensity data within the effective length to obtain vibration signal data along the cable. Utilizing the optical fiber sensing cable as a transmission medium, it senses the surrounding environment, enabling continuous monitoring of vibration signals over a range of hundreds of kilometers on a single optical fiber. The longer the optical fiber sensing cable, the weaker the light signal energy received by the optical module connected to it. The effective length of the optical fiber sensing cable is defined as the length of the location where the optical module can reach a set threshold of light intensity data.

[0053] The signal energy feature analysis module is used to analyze the energy feature values ​​of vibration signal data. This module includes a time-domain peak energy feature analysis unit, a signal envelope energy feature analysis unit, and a main frequency band energy feature analysis unit. The time-domain peak energy feature analysis unit calculates the time-domain peak energy feature value corresponding to the signal detected by each sensing unit; the signal envelope energy feature analysis unit calculates the signal envelope energy feature value corresponding to the signal detected by each sensing unit; and the main frequency band energy feature analysis unit calculates the main frequency band energy feature value corresponding to the signal detected by each sensing unit. Because the time-domain peak energy feature value, signal envelope energy feature value, and main frequency band energy feature value are complementary, comprehensively analyzing the target signal from multiple dimensions and multiple features can significantly improve the accuracy and robustness of the identification.

[0054] The adaptive dynamic adjustment module for each energy characteristic threshold parameter is used to adaptively and flexibly configure the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for each sensing unit. Through the adaptive dynamic adjustment technology of energy threshold parameters, the characteristic energy parameter thresholds corresponding to different sensing units are adaptively set. At the same time, it can flexibly adjust the thresholds at abnormal sensing unit points or junctions between different scenarios, further optimizing the system threshold parameter configuration and improving the system's practicality and robustness in complex scenarios.

[0055] The multi-feature energy threshold disturbance discrimination module is used to judge external disturbance events by combining the time-domain peak energy features, signal envelope energy features, and main frequency band energy features of each sensing unit, and to determine whether it is an external disturbance event. It compares the external signals detected by each sensing unit with the time-domain peak energy feature threshold, signal envelope energy feature threshold, and main frequency band energy feature threshold corresponding to each sensing unit to perform disturbance discrimination, and determines whether the external signals detected by each sensing unit are external vibration signals.

[0056] The vibration event pattern recognition module is used to perform pattern recognition on vibration signals identified as external disturbance events. By calculating the original signal waveform data identified as external vibration signals and their corresponding time-domain peak energy characteristic values, signal envelope energy characteristic values, and main frequency band energy characteristic values, the type of external vibration signal can be determined, which is the event type that generated the external vibration signal. The division of modules in the above-described distributed fiber optic vibration intelligent monitoring system adapted to multiple scenarios is only for illustrative purposes. In other embodiments, the distributed fiber optic vibration intelligent monitoring system adapted to multiple scenarios can be divided into different modules as needed to complete all or part of the functions of the above system.

[0057] In the description of this application, it should be noted that the terms "upper," "lower," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two elements. For those skilled in the art, the specific meaning of the above terms in this application can be understood according to the specific circumstances.

[0058] It should be noted that in this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0059] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A distributed fiber optic vibration intelligent monitoring method adaptable to multiple scenarios, characterized in that, The method includes the following steps: The light intensity data of the backscattered Rayleigh light signal in the optical fiber caused by the vibration source along the detection sensing optical cable; The effective length of the sensing optical cable is determined based on the light intensity data. The light intensity data within the effective length is demodulated to obtain the vibration signal data along the sensing optical cable. The energy characteristic values ​​of the vibration signal data are analyzed, including the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit. The time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold of each sensing unit are adaptively and flexibly configured dynamically. The external disturbance event is identified by combining the time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics to determine whether it is an external disturbance event signal. The vibration signal type identified as an external disturbance event will be subjected to pattern recognition. The effective length of the sensing optical cable is determined based on the light intensity data, specifically including: Set light intensity threshold I th ; The backscattered Rayleigh light signals detected at each monitoring point within the time interval t are: ; The light intensity information of each monitoring point was calculated according to Formula 1. X 1, X 2,…, X n ]; Formula 1 in: x ij Representing the i Time of the first j The light intensity values ​​corresponding to each monitoring point; Compare the light intensity values ​​at each monitoring point with the light intensity threshold. I th By comparison, the first signal in this set of data that exceeds the light intensity threshold is identified. I th The data value is denoted as X p And index the last one that is greater than the light intensity threshold. I th The data value is denoted as X q The starting point of the sensing optical cable can be calculated according to Formula 2. L 1 and termination point L 2: in: Fs The sampling rate of the data acquisition card. n f The refractive index coefficient of the sensing optical cable. c At the speed of light, p The first one greater than the light intensity threshold I th Number of data points q The last one is greater than the light intensity threshold. I th Number of data points; The effective length of the sensing optical cable is Intensity demodulation technology is used to demodulate the intensity of each data point within the effective length of the sensing optical cable, and finally the vibration signal data corresponding to each sensing unit is obtained. The time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold of each sensing unit are adaptively and flexibly configured dynamically, specifically including: Based on the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit, the corresponding time-domain peak energy characteristic curve of each sensing unit is obtained. G P Signal envelope energy characteristic curve G S and the main frequency band energy characteristic curve G F ; Polynomial curve fitting is performed on each energy characteristic curve using Formula 9 to achieve adaptive setting of the domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold parameters corresponding to each sensing unit. Set the corresponding adjustment curvature scaling factor and curve bias factor for the fitted curve respectively; Based on the formula, fit the time-domain peak energy characteristic curve. Signal envelope energy characteristic fitting curve Fitting curve of energy characteristics of the main frequency band Make dynamic adjustments; Also includes: Based on real-time data and changes in the current scene environment, flexible single-point threshold corrections can be performed on the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for certain abnormal sensing unit points or junctions between different scenes, resulting in optimized time-domain peak energy characteristic thresholds for each monitoring point. Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold ; Using the time-domain peak energy characteristic threshold of each sensing unit Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Disturbance discrimination is performed on the external signals detected at each monitoring point; The external disturbance event is determined by combining the time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics. Specifically, this includes: Thresholds of peak energy characteristics in the time domain Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Set the corresponding weight parameters The disturbance discrimination formula eleven is as follows: The disturbance event discrimination formula is based on the comprehensive information of time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics. in, The first i The time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of the signals detected at each monitoring point; If the time-domain peak energy characteristic value of the current target signal P Signal envelope energy eigenvalues S and the energy characteristic value of the main frequency band F At the same time greater than If the condition is met, the signal is identified as an abnormal disturbance event; otherwise, the signal is identified as a non-abnormal disturbance event.

2. The distributed optical fiber vibration intelligent monitoring method adaptable to multiple scenarios as described in claim 1, characterized in that, The time-domain peak energy characteristic values ​​obtained from analyzing vibration signal data include: Perform time-domain segmentation on the target signal, dividing the original time-domain signal into... m Each time domain segment is configured with a corresponding time domain peak-finding width. d 1, d 2,…, d m and peak-seeking threshold [ h 1, h 2,…, h m ]; Peak finding is performed in the positive and negative domains of the data amplitude for each time domain segment, where the first peak is... i The set of peaks found in the positive value region of each time domain segment is denoted as... Next, take the negative value of the target signal and perform peak finding in its negative range. The set of peaks found in the negative range is denoted as... ; Calculate the first [formula] using Formula 3. i The median of the peak set found for each time domain segment is denoted as the peak energy corresponding to that time domain segment. ; Formula 3 Formula 4 Finally, this can be calculated using Formula 4. m The average of the peak energies of each time-domain segment is denoted as the time-domain peak energy characteristic value corresponding to the target signal. P .

3. The distributed optical fiber vibration intelligent monitoring method adaptable to multiple scenarios as described in claim 1, characterized in that, The signal envelope energy characteristic values ​​obtained from analyzing vibration signal data include: Set the width of the time-domain sliding window. d The sliding window moves one step at a time. l Assuming the target signal data is [ y 1, y 2,…, y N According to Formula 5, the energy value of each sliding window is calculated by energy integration, and the integral energy curve is obtained and denoted as follows. ; Linear spline interpolation is used to perform interval linear interpolation on the data points in the integral energy curve. The number of data points interpolated is... The number of data points in the interpolated integral energy curve is the same as the number of data points in the original target signal, denoted as... ; Based on the calculated time-domain peak energy characteristic value P Set the threshold for the integral energy curve. P * d ,in: d As the weighting coefficient, sequentially with the threshold of the integral energy curve P * d Compare and find the data segment with the longest continuous value greater than the threshold of the integral energy curve, denoted as . ; Formula Six Finally, the signal envelope energy characteristic value S corresponding to the target signal is calculated according to Formula 6.

4. The distributed optical fiber vibration intelligent monitoring method adaptable to multiple scenarios as described in claim 1, characterized in that, The energy characteristic values ​​of the main frequency band obtained from the analysis of vibration signal data include: The spectral information corresponding to the target signal data is obtained through Fourier transform, and the maximum value of the signal spectral energy is found. Y max ; Set the main frequency band energy characteristic discrimination parameters a * Y max The signal spectrum data is sequentially compared with the main frequency band energy characteristic discrimination parameters. a * Y max Compare; Find all signals in the spectrum corresponding to this signal that are greater than [a certain value]. a * Y max The frequency value point corresponding to the frequency domain energy is denoted as . ; Find the maximum frequency point according to Formula 7. With minimum frequency point The difference is taken as the dominant frequency band of the signal: Formula 7 Formula 8 Then, by integrating all spectral energy values ​​in the main frequency band of the signal, the energy characteristic value of the main frequency band corresponding to the signal is obtained according to Formula 8. F .

5. The distributed optical fiber vibration intelligent monitoring method adaptable to multiple scenarios as described in claim 1, characterized in that, The vibration signal types identified as external disturbance events will be subjected to pattern recognition, specifically including: The event type of the disturbance event is identified by using the original signal waveform data corresponding to the identified disturbance event and its corresponding time-domain peak energy feature value, signal envelope energy feature value and main frequency band energy feature value. A neural network model is built to perform convolution calculations on the original signal waveform data corresponding to the disturbance event and extract the corresponding waveform features; The waveform features, along with the extracted time-domain peak energy features, signal envelope energy features, and main frequency band energy features, form a multi-dimensional feature matrix. This matrix is ​​then fed into a pattern classifier for matrix calculation to classify and identify the types of disturbance events.

6. A distributed fiber optic vibration intelligent monitoring system adaptable to multiple scenarios, characterized in that, include: A distributed optical fiber vibration detection module is used to detect the light intensity data of the backscattered Rayleigh light signal in the optical fiber caused by the vibration source along the sensing optical cable. The optical fiber sensing cable length calibration module is used to calibrate the effective length of the optical fiber sensing cable based on light intensity data, demodulate the light intensity data within the effective length, and obtain vibration signal data along the optical fiber sensing cable. The signal energy characteristic analysis module is used to analyze the energy characteristic values ​​of vibration signal data. The energy characteristic values ​​include the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit. An adaptive dynamic adjustment module for each energy characteristic threshold parameter is used to adaptively and flexibly configure the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for each sensing unit. Based on the multi-feature energy threshold disturbance discrimination module, it is used to judge external disturbance events by combining the time-domain peak energy features, signal envelope energy features and main frequency band energy features of each sensing unit, and to determine whether it is an external disturbance event signal. The vibration event pattern recognition module is used to perform pattern recognition on the vibration signal type that is determined to be an external disturbance event. The effective length of the sensing optical cable is determined based on the light intensity data, specifically including: Set light intensity threshold I th ; The backscattered Rayleigh light signals detected at each monitoring point within the time interval t are: ; The light intensity information of each monitoring point was calculated according to Formula 1. X 1, X 2,…, X n ]; Formula 1 in: x ij Representing the i Time of the first j The light intensity values ​​corresponding to each monitoring point; Compare the light intensity values ​​at each monitoring point with the light intensity threshold. I th By comparison, the first signal in this set of data that exceeds the light intensity threshold is identified. I th The data value is denoted as X p And index the last one that is greater than the light intensity threshold. I th The data value is denoted as X q The starting point of the sensing optical cable can be calculated according to Formula 2. L 1 and termination point L 2: in: Fs The sampling rate of the data acquisition card. n f The refractive index coefficient of the sensing optical cable. c At the speed of light, p The first one greater than the light intensity threshold I th Number of data points q The last one is greater than the light intensity threshold. I th Number of data points; The effective length of the sensing optical cable is Intensity demodulation technology is used to demodulate the intensity of each data point within the effective length of the sensing optical cable, and finally the vibration signal data corresponding to each sensing unit is obtained. The time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold of each sensing unit are adaptively and flexibly configured dynamically, specifically including: Based on the time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of each sensing unit, the corresponding time-domain peak energy characteristic curve of each sensing unit is obtained. G P Signal envelope energy characteristic curve G S and the main frequency band energy characteristic curve G F ; Polynomial curve fitting is performed on each energy characteristic curve using Formula 9 to achieve adaptive setting of the domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold parameters corresponding to each sensing unit. Set the corresponding adjustment curvature scaling factor and curve bias factor for the fitted curve respectively; Based on the formula, fit the time-domain peak energy characteristic curve. Signal envelope energy characteristic fitting curve Fitting curve of energy characteristics of the main frequency band Make dynamic adjustments; Also includes: Based on real-time data and changes in the current scene environment, flexible single-point threshold corrections can be performed on the time-domain peak energy characteristic threshold, signal envelope energy characteristic threshold, and main frequency band energy characteristic threshold for certain abnormal sensing unit points or junctions between different scenes, resulting in optimized time-domain peak energy characteristic thresholds for each monitoring point. Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold ; Using the time-domain peak energy characteristic threshold of each sensing unit Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Disturbance discrimination is performed on the external signals detected at each monitoring point; The external disturbance event is determined by combining the time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics. Specifically, this includes: Thresholds of peak energy characteristics in the time domain Signal envelope energy characteristic threshold and the main frequency band energy characteristic threshold Set the corresponding weight parameters The disturbance discrimination formula eleven is as follows: The disturbance event discrimination formula is based on the comprehensive information of time-domain peak energy characteristics, signal envelope energy characteristics, and main frequency band energy characteristics. in, The first i The time-domain peak energy characteristic value, signal envelope energy characteristic value, and main frequency band energy characteristic value of the signals detected at each monitoring point; If the time-domain peak energy characteristic value of the current target signal P Signal envelope energy eigenvalues S and the energy characteristic value of the main frequency band F At the same time greater than If the condition is met, the signal is identified as an abnormal disturbance event; otherwise, the signal is identified as a non-abnormal disturbance event.

Images