A key method and system for pipeline monitoring risk warning in diversified application scenarios

Abstract

The present invention discloses a key method and system for early warning of pipeline monitoring risks in diversified application scenarios, including the following steps: S1, collecting temperature sensor data and vibration sensor data; S2, correlating and analyzing the collected data, and performing machine learning for event recognition based on the collected data to obtain a mixed Gaussian model; S3, performing scene recognition based on the mixed Gaussian model, judging the type of pipeline abnormality and issuing an early warning. Accurate judgment of pipeline events and risk decision-making early warning in diversified application scenarios including vibration alarm, fire warning, and submarine cable external damage prevention are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline monitoring, and in particular to a key method and system for early warning of pipeline monitoring risks in diversified application scenarios. Background Art

[0002] The pipeline monitoring and early warning system is used in power systems and other units with massive line management needs. The pipeline intelligent perception system developed using fiber optic sensing technology and big data analysis technology proposes a pipeline event identification and early warning method, which combines the basic data such as the time of occurrence, geographical location, and disturbance type of pipeline line events with the inherent characteristics of various harmful events in power pipelines in the above basic data dimensions, and uses the correlation algorithm of time, space, and type to determine the full picture of the pipeline event on site.

[0003] The existing technology can only identify the event type through vibration sensing and sound sensing. The main event type identified is construction-related external damage prevention. There is no monitoring and identification mechanism for fire events that may be caused by overheating of joints caused by long-term operation of cables in pipelines. Compared with land cables, power cables are also laid at a certain depth on the seabed. At the same time, submarine cables also have different fault problems from land cables. Usually, cables passing through ports are easily damaged by anchors of surrounding ships. The main event type identified by the existing system is construction-related external damage prevention within the land range, and there is no effective monitoring of external damage to submarine cables. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a key method and system for pipeline monitoring risk warning in diversified application scenarios, combine temperature sensor information, analyze power cable pipeline events, and achieve more comprehensive and accurate monitoring and alarm of fault types.

[0005] In order to solve the above technical problems, a technical solution adopted by the present invention is:

[0006] A key method for pipeline monitoring risk warning in diversified application scenarios includes the following steps:

[0007] S1, collecting temperature sensor data and vibration sensor data;

[0008] S2, correlation analysis of the collected data, and machine learning for event recognition based on the collected data to obtain a mixed Gaussian model;

[0009] S3. Perform scene recognition based on the mixed Gaussian model to determine the type of pipeline abnormality and issue an early warning.

[0010] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0011] A pipeline monitoring risk early warning system, comprising:

[0012] one or more processors;

[0013] A storage device for storing one or more programs;

[0014] When the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned key method for risk warning of pipeline monitoring in diversified application scenarios.

[0015] The beneficial effects of the present invention are: providing a key method and system for pipeline monitoring risk warning in diversified application scenarios, combining vibration sensing and temperature sensing information to judge power cable pipeline events, and realize monitoring and alarm of fault types such as overheating, discharge, and fire. No sensors need to be installed along the entire length of the pipeline, and signal monitoring and sensing can be achieved through ordinary optical fibers. The overall realization includes accurate judgment and risk warning of pipeline events in diversified application scenarios such as vibration alarm, fire warning, and submarine cable external damage prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of a key method for early warning of pipeline monitoring risks in diversified application scenarios according to an embodiment of the present invention;

[0017] Figure 2 A schematic diagram of a pipeline monitoring risk early warning system for diversified application scenarios according to an embodiment of the present invention;

[0018] Figure 3 It is an event recognition machine learning algorithm model of a key method for early warning of pipeline monitoring risks in diversified application scenarios of an embodiment of the present invention;

[0019] Figure 4 This is a flow chart of tidal characteristic signal processing of a key method for risk warning of pipeline monitoring in diversified application scenarios of an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0021] A key method for pipeline monitoring risk warning in diversified application scenarios includes the following steps:

[0022] S1, collecting temperature sensor data and vibration sensor data;

[0023] S2, correlation analysis of the collected data, and machine learning for event recognition based on the collected data to obtain a mixed Gaussian model;

[0024] S3. Perform scene recognition based on the mixed Gaussian model to determine the type of pipeline abnormality and issue an early warning.

[0025] From the above description, it can be seen that the beneficial effects of the present invention are: providing a key method for pipeline monitoring risk warning in diversified application scenarios, combining vibration sensing and temperature sensing information, analyzing power cable pipeline events, and realizing monitoring and alarming of fault types such as overheating, discharge, and fire. No sensors need to be installed along the entire length of the pipeline, and signal monitoring and sensing can be achieved through ordinary optical fibers. The overall realization includes accurate analysis and risk warning of pipeline events in diversified application scenarios such as vibration alarm, fire warning, and submarine cable external damage prevention.

[0026] Furthermore, the machine learning for event recognition based on the data in step S2 is specifically as follows:

[0027] S201, construct the initial mixed Gaussian model as:

[0028] Among them, X is a random variable, N is the number of data, X is a random variable, μ k is the mean, σ k 2 is the variance, and k is the total number of clusters.

[0029] N(Xμ k , σ k 2 ) is the kth cluster in the Gaussian mixture model, and the mixing coefficient Π k is the weight of each single Gaussian model, satisfying Π k >0,

[0030] S202, calculation data X n The probability of belonging to the kth cluster:

[0031]

[0032] When there are two dimensions, vibration dimension and temperature dimension:

[0033]

[0034]

[0035] The mean μ is estimated using an iterative method k and variance σ k 2 The value of the previous iteration is taken as the initial value;

[0036] S203, repeat the calculation step of S202 until the parameter values ​​of the mixed Gaussian model are stable.

[0037] From the above description, it can be seen that the same line is monitored by combining vibration sensing and temperature sensing. The data collected by temperature sensing is combined with the data collected by vibration sensing. The two provide a new reference dimension for event analysis through data correlation analysis. In the pipeline, the geographical location of the event and the intensity of the disturbance are determined by vibration to determine whether it needs to be handled on site. At the same time, combined with the temperature abnormality data of the temperature sensor, it is confirmed whether the pipeline has been damaged, resulting in leakage and other faults.

[0038] Furthermore, the data collected in step S2 also includes a tidal characteristic signal, and the processing steps for the tidal characteristic signal are specifically as follows:

[0039] The ocean tide characteristic signal is approximated by adding a fixed modulation acos(ω0t) to the sensing light signal;

[0040] The sensing light signal is light emitted by the same light source divided into a first light beam and a second light beam, the first light beam is subjected to a delay process, and the time difference between the first light beam and the second light beam is a constant 2L / C+τ;

[0041] According to the fixed modulation, at time t, the first light beam is modulated by the tide to acos(ω0t);

[0042] According to the fixed modulation, At the moment, the second light beam is modulated by the tide to

[0043] The phase difference calculation formulas for the influence of the ocean tide characteristic signal on the first light beam and the second light beam are established in turn:

[0044]

[0045] in, is the initial phase, φ0 is the phase after modulation, n is the refractive index of the optical fiber, L is the optical path length, c is the speed of light, τ is the delay constant, and ω0 is the optimal value obtained by measuring the optical fiber sensing system in a tidal environment without other interference events.

[0046] From the above description, it can be seen that when there is a submarine cable section in the line monitored by the system, the error caused by the influence of the sea tide can be reduced in positioning and event identification. In view of the working scene of the submarine cable, the interference signal in the sea area is processed, so as to effectively and accurately monitor the external damage of the submarine cable. A more comprehensive grasp of the information of the land cable and submarine cable events is achieved, and the early warning capability of harmful events is improved.

[0047] Furthermore, determining the type of pipeline abnormality in step S3 includes:

[0048] If only the vibration sensor sends abnormal data, the abnormal point of the pipeline is located and determined to be affected by external construction;

[0049] If only the temperature sensor sends out abnormal temperature rise data, the abnormal point of the pipeline is located to determine whether the temperature rise reaches the abnormal threshold. If it reaches the abnormal threshold, it is determined to be local overheating.

[0050] From the above description, it can be seen that by using the parameter changes in a single dimension of temperature sensing or vibration sensing, it can be confirmed whether the cable duct has been locally overheated or affected by external construction.

[0051] Furthermore, determining the type of pipeline abnormality in step S3 further includes:

[0052] If both the vibration sensor and the temperature sensor send out abnormal data, the abnormal point of the pipeline is located, and whether there is discharge is determined by the vibration data type; if there is a discharge condition, it is judged as local discharge, otherwise it is judged as a fire.

[0053] From the above description, it can be seen that by combining the information of vibration sensing and temperature sensing, power cable pipeline events can be analyzed and judged to achieve monitoring of fault types such as overheating, discharge, and fire.

[0054] A pipeline monitoring risk early warning system for diversified application scenarios includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0055] S1, collecting temperature sensor data and vibration sensor data;

[0056] S2, correlation analysis of the collected data, and machine learning for event recognition based on the collected data to obtain a mixed Gaussian model;

[0057] S3. Perform scene recognition based on the mixed Gaussian model to determine the type of pipeline abnormality and issue an early warning.

[0058] From the above description, it can be seen that the beneficial effects of the present invention are: providing a pipeline monitoring risk warning system for diversified application scenarios, combining vibration sensing and temperature sensing information, analyzing power cable pipeline events, and realizing monitoring and alarming of fault types such as overheating, discharge, and fire. No sensors need to be installed along the entire length of the pipeline, and signal monitoring and sensing can be realized through ordinary optical fibers. Accurate analysis and risk warning of pipeline events in diversified application scenarios including vibration alarms, fire warnings, and submarine cable external damage prevention are realized.

[0059] Furthermore, the machine learning for event recognition based on the data in step S2 is specifically as follows:

[0060] S201, construct the initial mixed Gaussian model as:

[0061] Among them, X is a random variable, N is the number of data, X is a random variable, μ k is the mean, σ k 2 is the variance, and k is the total number of clusters.

[0062] N(Xμ k , σ k 2 ) is the kth cluster in the Gaussian mixture model, and the mixing coefficient Π k is the weight of each single Gaussian model, satisfying Π k >0,

[0063] S202, calculation data X n The probability of belonging to the kth cluster:

[0064]

[0065] When there are two dimensions, vibration dimension and temperature dimension:

[0066]

[0067]

[0068] The mean μ is estimated using an iterative method k and variance σ k 2 The value of the previous iteration is taken as the initial value;

[0069] S203, repeat the calculation step of S202 until the parameter values ​​of the mixed Gaussian model are stable.

[0070] From the above description, it can be seen that the same line is monitored by combining vibration sensing and temperature sensing. The data collected by temperature sensing is combined with the data collected by vibration sensing. The two provide a new reference dimension for event analysis through data correlation analysis. In the pipeline, the geographical location of the event and the intensity of the disturbance are determined by vibration to determine whether it needs to be handled on site. At the same time, combined with the temperature abnormality data of the temperature sensor, it is confirmed whether the pipeline has been damaged, resulting in leakage and other faults.

[0071] Furthermore, the data collected in step S2 also includes a tidal characteristic signal, and the processing steps for the tidal characteristic signal are specifically as follows:

[0072] The ocean tide characteristic signal is approximated by adding a fixed modulation acos(ω0t) to the sensing light signal;

[0073] The sensing light signal is light emitted by the same light source divided into a first light beam and a second light beam, the first light beam is subjected to a delay process, and the time difference between the first light beam and the second light beam is a constant 2L / C+τ;

[0074] According to the fixed modulation, at time t, the first light beam is modulated by the tide to acos(ω0t);

[0075] According to the fixed modulation, At the moment, the second light beam is modulated by the tide to

[0076] The phase difference calculation formulas for the influence of the ocean tide characteristic signal on the first light beam and the second light beam are established in turn:

[0077]

[0078] in, is the initial phase, φ0 is the phase after modulation, n is the refractive index of the optical fiber, L is the optical path length, c is the speed of light, τ is the delay constant, and ω0 is the optimal value obtained by measuring the optical fiber sensing system in a tidal environment without other interference events.

[0079] From the above description, it can be seen that when there is a submarine cable section in the line monitored by the system, the error caused by the influence of the sea tide can be reduced in positioning and event identification. In view of the working scene of the submarine cable, the interference signal in the sea area is processed, so as to effectively and accurately monitor the external damage of the submarine cable. A more comprehensive grasp of the information of the land cable and submarine cable events is achieved, and the early warning capability of harmful events is improved.

[0080] Furthermore, determining the type of pipeline abnormality in step S3 includes:

[0081] If only the vibration sensor sends abnormal data, the abnormal point of the pipeline is located and determined to be affected by external construction;

[0082] If only the temperature sensor sends out abnormal temperature rise data, the abnormal point of the pipeline is located to determine whether the temperature rise reaches the abnormal threshold. If it reaches the abnormal threshold, it is determined to be local overheating.

[0083] From the above description, it can be seen that by using the parameter changes in a single dimension of temperature sensing or vibration sensing, it can be confirmed whether the cable duct has been locally overheated or affected by external construction.

[0084] Furthermore, determining the type of pipeline abnormality in step S3 further includes:

[0085] If both the vibration sensor and the temperature sensor send out abnormal data, the abnormal point of the pipeline is located, and whether there is discharge is determined by the vibration data type; if there is a discharge condition, it is judged as local discharge, otherwise it is judged as a fire.

[0086] From the above description, it can be seen that by combining the information of vibration sensing and temperature sensing, power cable pipeline events can be analyzed and judged to achieve monitoring of fault types such as overheating, discharge, and fire.

[0087] The above-mentioned key method and system for pipeline monitoring risk warning in diversified application scenarios of the present invention can realize accurate judgment and risk warning of pipeline events in diversified application scenarios including vibration alarm, fire warning, submarine cable external damage prevention, etc., which are described below through specific implementation methods:

[0088] Embodiment 1

[0089] Please refer to Figure 1 , a key method for pipeline monitoring risk warning in diversified application scenarios, including the following steps:

[0090] S1, collecting temperature sensor data and vibration sensor data;

[0091] Specifically, vibration sensing and temperature sensing use distributed sensing with full-line perception. The data comes from the process of light propagating in the optical fiber. Certain parameters are changed due to the influence of external events. The system connected to the optical fiber terminal analyzes the changed parameters generated by the demodulated optical signal to obtain information related to the vibration or temperature change event.

[0092] S2, correlation analysis of the collected data, and machine learning for event recognition based on the collected data to obtain a mixed Gaussian model;

[0093] Please refer to Figure 3 In step S2, the machine learning for event recognition based on the data is specifically as follows:

[0094] S201, construct the initial mixed Gaussian model as:

[0095] Among them, X is a random variable, N is the number of data, X, μ k is the mean, σ k 2 is the variance, and k is the total number of clusters.

[0096] N(Xμ k , σ k 2 ) is the kth cluster in the Gaussian mixture model, and the mixing coefficient Π k is the weight of each single Gaussian model, satisfying Π k >0,

[0097] S202, calculation data X n The probability of belonging to the kth cluster:

[0098]

[0099] When there is only the vibrational dimension:

[0100] When there are two dimensions, vibration dimension and temperature dimension:

[0101]

[0102]

[0103] The mean μ is estimated using an iterative method k and variance σ k 2 The value of the previous iteration is taken as the initial value;

[0104] S203, repeat the calculation step of S202 until the parameter values ​​of the mixed Gaussian model are stable.

[0105] The improved algorithm can simultaneously associate vibration and temperature related samples for training and learning, so that in the subsequent process the system can automatically identify the two types of information, thereby increasing the richness of line event recognition.

[0106] S3. Perform scene recognition based on the mixed Gaussian model to determine the type of pipeline abnormality and issue an early warning.

[0107] This embodiment solves the key problem of machine learning after adding the temperature dimension, making it feasible to accurately identify events under the combination of vibration and temperature dimensions.

[0108] Embodiment 2

[0109] Please refer to Figure 4 The difference between this embodiment and the first embodiment is that, for monitoring the submarine cable laid in the sea environment, the characteristic signal of the sea tide is also collected, and the influence of the sea tide is reduced through algorithm processing, so that the pipeline monitoring data is more accurate. The sea tide is a harmless interference, which is different from the interference of road vehicle movement, natural weather, etc. Therefore, the specific processing mechanism in this embodiment is used to reduce the noise of the signal influence caused by the sea tide.

[0110] Specifically, the data collected in step S2 also includes a tidal characteristic signal, and the steps for processing the tidal characteristic signal are:

[0111] The ocean tide characteristic signal is approximated by adding a fixed modulation acos(ω0t) to the sensor light signal;

[0112] The sensing light signal is the light emitted by the same light source divided into a first light beam and a second light beam. The first light beam is delayed, and the time difference between the first light beam and the second light beam is a constant 2L / C+τ.

[0113] According to the fixed modulation, at time t, the first beam is modulated by the tide to acos(ω0t);

[0114] According to the fixed modulation, At time t

[0115] The phase difference calculation formulas for the influence of the ocean tide characteristic signal on the first light beam and the second light beam are established in turn:

[0116]

[0117] in, is the initial phase, φ0 is the phase after modulation, n is the refractive index of the optical fiber, L is the optical path length, c is the speed of light, τ is the delay constant, and ω0 is the optimal value obtained by measuring the optical fiber sensing system in a tidal environment without other interference events.

[0118] This embodiment solves the problem of determining the fixed interference caused by the sea tide on the optical fiber sensing system, so that the positioning and pattern recognition of the optical fiber sensing in the sea tide environment can obtain accurate values, thereby enhancing the practicality of the system in environments such as sea tides.

[0119] Embodiment 3

[0120] The difference between this embodiment and the first or second embodiment is that the basis for determining the type of abnormal condition of the pipeline is further limited:

[0121] If only the vibration sensor sends abnormal data, the abnormal point of the pipeline is located and determined to be affected by external construction;

[0122] Specifically, the geographical location of the event is determined by vibration in the power pipeline, and whether on-site processing is required is determined based on the intensity of the disturbance.

[0123] If only the temperature sensor sends out abnormal temperature rise data, the abnormal point of the pipeline is located to determine whether the temperature rise reaches the abnormal threshold. If it reaches the abnormal threshold, it is determined to be local overheating.

[0124] Specifically, the impact of hazardous events can be determined through sudden changes in temperature; the continuous state of the corresponding hazardous event can be determined through continuous changes in temperature, such as: ongoing, recovered; a continuous temperature change curve within a certain line range is provided to accurately determine the start time and duration of the impact of the hazardous event.

[0125] If both the vibration sensor and the temperature sensor send out abnormal data, the abnormal point of the pipeline is located, and whether there is discharge is determined by the vibration data type; if there is a discharge condition, it is judged as local discharge, otherwise it is judged as a fire.

[0126] Specifically, the correlation and effective processing mechanism of the same monitoring line events in the two dimensions of vibration sensing and temperature sensing are realized. On the basis of the existing monitoring system, the alarm event types of overheating and fire warning are added. It is proposed to realize the correlation and effective processing mechanism of the same monitoring line events in the two dimensions of vibration sensing and temperature sensing based on Sagnac interference and Raman effect.

[0127] Embodiment 4

[0128] Please refer to Figure 2 A pipeline monitoring risk warning system for diversified application scenarios includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in embodiments one to three are implemented.

[0129] In summary, the present invention provides a key method and system for pipeline monitoring risk warning in diversified application scenarios. By combining vibration sensing and temperature sensing information, power cable pipeline events are analyzed and judged, thereby realizing accurate analysis and risk decision-making warning of pipeline events in diversified application scenarios including vibration alarm, fire warning, and submarine cable external damage prevention.

[0130] It should be noted that, for the convenience of description, the aforementioned method embodiments are all described as a series of action combinations, but those skilled in the art should be aware that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0131] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0132] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A key method for pipeline monitoring risk warning in diversified application scenarios, characterized in that: The following steps are involved: S1, collecting temperature sensor data and vibration sensor data; S2. Correlate and analyze the collected data, and perform machine learning for event recognition based on the collected data to obtain a mixed Gaussian model, specifically: S201, construct the initial mixed Gaussian model as: Among them, X is a random variable, N is the number of data, μ k is the mean, is the variance, k is the total number of clusters; is the kth cluster in the Gaussian mixture model, and the mixing coefficient Π k is the weight of each single Gaussian model, satisfying Π k >0, S202, calculation data X n The probability of belonging to the kth cluster: When there are two dimensions, vibration dimension and temperature dimension: The mean μ is estimated using an iterative method k and The value of the previous iteration is taken as the initial value; S203, repeat the calculation step of S202 until the parameter value of the mixed Gaussian model is stable; The collected data also includes a tidal characteristic signal, and the processing steps for the tidal characteristic signal are specifically as follows: The ocean tide characteristic signal is approximated by adding a fixed modulation acos(ω0t) to the sensing light signal; The sensing light signal is light emitted by the same light source divided into a first light beam and a second light beam, the first light beam is subjected to a time delay process, and the time difference between the first light beam and the second light beam is a constant 2L / c+τ; According to the fixed modulation, at time t, the first light beam is modulated by the tide to acos(ω0t); According to the fixed modulation, At the moment, the second light beam is modulated by the tide to The phase difference calculation formulas for the influence of the ocean tide characteristic signal on the first light beam and the second light beam are established in turn: in, is the initial phase, φ0 is the phase after modulation, n is the refractive index of the optical fiber, L is the optical path length, c is the speed of light, τ is the delay constant, and ω0 is the optimal value obtained by measuring the optical fiber sensing system in a tidal environment without other interference events; S3. Perform scene recognition based on the mixed Gaussian model to determine the type of pipeline abnormality and issue an early warning.

2. According to a key method for early warning of pipeline monitoring risks in diversified application scenarios according to claim 1, it is characterized in that: Determining the type of pipeline abnormality in step S3 includes: If only the vibration sensor sends abnormal data, the abnormal point of the pipeline is located and determined to be affected by external construction; If only the temperature sensor sends out abnormal temperature rise data, the abnormal point of the pipeline is located to determine whether the temperature rise reaches the abnormal threshold. If it reaches the abnormal threshold, it is determined to be local overheating.

3. According to a key method for early warning of pipeline monitoring risks in diversified application scenarios according to claim 1, it is characterized in that: Determining the type of pipeline abnormality in step S3 also includes: If both the vibration sensor and the temperature sensor send out abnormal data, the abnormal point of the pipeline is located, and whether there is discharge is determined by the vibration data type; if there is a discharge condition, it is judged as local discharge, otherwise it is judged as a fire.

4. A pipeline monitoring risk warning system with diversified application scenarios, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the following steps are implemented: S1, collecting temperature sensor data and vibration sensor data; S2. Correlate and analyze the collected data, and perform machine learning for event recognition based on the collected data to obtain a mixed Gaussian model, specifically: S201, construct the initial mixed Gaussian model as: Among them, X is a random variable, N is the number of data, μ k is the mean, is the variance, k is the total number of clusters; is the kth cluster in the Gaussian mixture model, and the mixing coefficient Π k is the weight of each single Gaussian model, satisfying ∏ k >0, S202, calculation data X n The probability of belonging to the kth cluster: When there are two dimensions, vibration dimension and temperature dimension: The mean μ is estimated using an iterative method k and The value of the previous iteration is taken as the initial value; S203, repeat the calculation step of S202 until the parameter value of the mixed Gaussian model is stable; The collected data also includes a tidal characteristic signal, and the processing steps for the tidal characteristic signal are specifically as follows: The ocean tide characteristic signal is approximated by adding a fixed modulation acos(ω0t) to the sensing light signal; The sensing light signal is light emitted by the same light source divided into a first light beam and a second light beam, the first light beam is subjected to a delay process, and the time difference between the first light beam and the second light beam is a constant 2L / C+τ; According to the fixed modulation, at time t, the first light beam is modulated by the tide to acos(ω0t); According to the fixed modulation, At the moment, the second light beam is modulated by the tide to The phase difference calculation formulas for the influence of the ocean tide characteristic signal on the first light beam and the second light beam are established in turn: in, is the initial phase, φ0 is the phase after modulation, n is the refractive index of the optical fiber, L is the optical path length, c is the speed of light, τ is the delay constant, and ω0 is the optimal value obtained by measuring the optical fiber sensing system in a tidal environment without other interference events; S3. Perform scene recognition based on the mixed Gaussian model to determine the type of pipeline abnormality and issue an early warning.

5. According to the diversified application scenario pipeline monitoring risk early warning system of claim 4, it is characterized in that: Determining the type of pipeline abnormality in step S3 includes: If only the vibration sensor sends abnormal data, the abnormal point of the pipeline is located and determined to be affected by external construction; If only the temperature sensor sends out abnormal temperature rise data, the abnormal point of the pipeline is located to determine whether the temperature rise reaches the abnormal threshold. If it reaches the abnormal threshold, it is determined to be local overheating.

6. According to the diversified application scenario pipeline monitoring risk warning system of claim 4, it is characterized in that: Determining the type of pipeline abnormality in step S3 also includes: If both the vibration sensor and the temperature sensor send out abnormal data, the abnormal point of the pipeline is located, and whether there is discharge is determined by the vibration data type; if there is a discharge condition, it is judged as local discharge, otherwise it is judged as a fire.

Images