Fire protection pipeline monitoring method and system based on optical fiber sensing and optical fiber

Abstract

The present invention relates to the field of optical fiber sensing technology, and specifically discloses a fire protection pipeline monitoring method and system based on optical fiber sensing, and an optical fiber, comprising: winding an optical fiber around a fire protection pipeline to be monitored, connecting one end of the optical fiber to a light source output end, and connecting the other end of the optical fiber to a vibration signal acquisition end; outputting signals of different frequency bands from the light source output end, and then acquiring the spectrum signal of the fire protection pipeline by the vibration signal acquisition end; normalizing the spectrum signal, mapping the amplitude data of the spectrum signal to between [0, 1]; performing wavelet packet decomposition on the normalized data using the db6 wavelet basis function and calculating the wavelet energy ratio of each wavelet packet node. The inverted triangle tip velvet structure in the present invention amplifies the mechanical vibration of a 0.1mm crack in the pipeline by more than three times through the geometric stress concentration effect, and the acoustic impedance matching characteristics of the nanofilling layer reduce the signal transmission loss by 40%, and can also preliminarily filter out redundant noise signals.

Description

Technical Field

[0001] The present invention relates to the field of optical fiber sensing technology, and specifically discloses a fire protection pipeline monitoring method and system based on optical fiber sensing, and an optical fiber. Background Art

[0002] Pipeline transportation boasts advantages such as high capacity, minimal land occupation, short construction periods, safety, reliability, continuity, and high efficiency. After years of development, it has become one of the five major modes of transportation, playing an irreplaceable role in improving people's living standards and national economic development. As a versatile resource transport device, pipelines transport vast quantities of fresh water, fuel, crude oil, and natural gas, and various types of pipelines have gradually become critical infrastructure.

[0003] Firefighting pipelines are used to connect firefighting equipment and devices, and to transport firefighting water, gas, or other media. Due to special requirements, firefighting pipelines have specific thickness and material requirements and are painted red. During pipeline operation, various pipeline anomalies are inevitable, including blockages, corrosion defects, and breakage and leaks.

[0004] Chinese patent CN118112713A discloses a thin-diameter, bend-resistant optical fiber for ocean monitoring, belonging to the field of optical fiber sensor technology. The invention comprises a fiber core, an inner cladding, a porous nanoring, an outer cladding surrounding the core, and a fiber Bragg grating (FBG) inscribed within the bend-resistant optical fiber. However, if the aforementioned patent discloses a technique directly used as a fiber sensor on a fire protection pipeline, the signals collected by the fiber sensor often contain excessive noise, lacking the ability to perform preliminary physical filtering of the noise.

[0005] The prior art includes a multi-risk early warning system for long tunnel fire protection pipelines based on fiber optic sensing. The Brillouin and Rayleigh light detection units utilize dual-wavelength lasers to implement combined detection of Brillouin and Rayleigh light on a fiber group. Combined with optical and electrical modulation devices, these units achieve joint detection of strain, vibration, and temperature signals. The fiber group, tightly attached to the fire protection pipeline, includes a tight fiber and a loose fiber. The tight fiber is used to detect strain signals in the fire protection pipeline, while the loose fiber is used to detect temperature and vibration signals in the fire protection pipeline. The fault determination unit determines the type of fault in the fire protection pipeline based on the output signal from the fiber group. In the prior art described above, different judgment thresholds cannot be set based on the type of pipeline abnormality, such as pipeline blockage or pipeline defect, and the effect of setting a preliminary warning or alarm based on the degree of blockage or defect cannot be achieved. Summary of the Invention

[0006] In view of the above deficiencies in the prior art, the present invention provides a fire pipeline monitoring method, system and optical fiber based on optical fiber sensing to solve the problems that traditional optical fiber sensing cannot filter out redundant noise signals and cannot set preliminary warnings or alarms according to different degrees of blockage or defects.

[0007] The technical solution adopted by the present invention to solve its technical problem is:

[0008] The fire protection pipeline monitoring method based on optical fiber sensing includes the following steps:

[0009] S1. Wind the optical fiber around the fire protection pipeline to be monitored, with one end of the optical fiber connected to the light source output end and the other end connected to the vibration signal acquisition end;

[0010] S2. The light source output terminal outputs signals of different frequency bands, and the vibration signal acquisition terminal collects the spectrum signal of the fire protection pipeline;

[0011] S3, performing normalization processing on the spectrum signal to map the amplitude data of the spectrum signal to the range [0, 1];

[0012] S4, using the db6 wavelet basis function to perform wavelet packet decomposition on the normalized data and calculate the wavelet energy ratio of each wavelet packet node;

[0013] S5. Generate a fire monitoring model based on the number of wavelet packet nodes of the maximum wavelet energy ratio, the maximum wavelet energy ratio, and the state relationship of the fire pipeline, and calculate the abnormality coefficient of the monitored fire pipeline through the fire monitoring model;

[0014] S6. Establish different judgment thresholds according to the abnormal coefficient and the state of the fire protection pipeline, and issue a fire alarm of the corresponding state when the abnormal coefficient of the monitored fire protection pipeline exceeds the judgment threshold.

[0015] Furthermore, in step S1, the number of turns of the optical fiber wound on the fire protection pipeline is not less than 30 turns per meter of the fire protection pipeline.

[0016] Furthermore, in step S2, the signals of different frequency bands are signals of any frequency band between 0-250 Hz.

[0017] Furthermore, in step S4, the signal is decomposed by wavelet packets and the energy proportion of each frequency band is calculated respectively. Since the total energy of the signal remains unchanged, the energy proportion of different frequency bands is:

[0018] ;in ;

[0019] Where N is the signal length, d i(k) is the value of the kth sampling point corresponding to the i-th frequency band after signal decomposition; E i represents the energy of the i-th frequency band; the total signal energy E is:

[0020] ;

[0021] The wavelet energy proportion m of the corresponding i-th frequency band i for:

[0022] .

[0023] Furthermore, in step S5, the specific steps of generating the fire monitoring model are:

[0024] S501. Evaluate the abnormality coefficient of the fire protection pipeline according to the degree of blockage and defect of the fire protection pipeline;

[0025] S502, establishing relational feature 1 based on the blockage degree of the fire protection pipeline and the number of wavelet packet nodes of the maximum wavelet energy ratio;

[0026] S503, establishing relationship feature 2 based on the blockage degree of the fire protection pipeline and the wavelet energy ratio;

[0027] S504: Based on the above relationship feature 1 and relationship feature 2, a fire monitoring model is fitted and generated.

[0028] Furthermore, the fire monitoring model is generated to calculate the abnormality coefficient of the monitored fire pipeline. The specific formula is:

[0029] ;

[0030] Where g represents the abnormal coefficient of the monitored fire pipeline; n represents the number of wavelet packet nodes with the maximum wavelet energy ratio; and m represents the maximum wavelet energy ratio.

[0031] Furthermore, when g<40%, it indicates that the fire protection pipeline is in normal condition; when 40%≤g≤60%, it indicates that the inner wall of the fire protection pipeline has a slight abnormality; and when g>60%, it indicates that the inner wall of the fire protection pipeline has a serious fault.

[0032] Furthermore, in step S6, the judgment thresholds for the blocked pipe and the pipe with inner wall defect are set based on the comparison of the spectrum signals and abnormal coefficients of the blocked pipe, the pipe with inner wall defect and the normal pipe.

[0033] The fire protection pipeline monitoring system based on optical fiber sensing includes:

[0034] The light source output module inputs the light source signal of any frequency band between 0-250Hz into the optical fiber;

[0035] The vibration signal acquisition module collects the spectrum signal output by the light source output module from the other end of the optical fiber;

[0036] The data processing module normalizes the spectrum information and uses wavelet packets to decompose and calculate the wavelet energy ratio of each wavelet packet node;

[0037] The feature analysis module generates a fire monitoring model based on the number of wavelet packet nodes of the maximum wavelet energy ratio, the relationship between the maximum wavelet energy ratio and the state of the fire pipeline, and calculates the abnormal coefficient of the monitored fire pipeline through the fire monitoring model;

[0038] The fire alarm module will sound an alarm when the abnormal coefficient of the monitored fire pipeline exceeds the set threshold.

[0039] The optical fiber for realizing the fire pipeline monitoring method based on optical fiber sensing includes a fiber core, an inner cladding, a tip fleece, a nano-filling layer and an outer cladding. Several tip fleeces are arranged inside the nano-filling layer, and several tip fleeces are located on the outer surface of the inner cladding. The cross-sectional profile of the tip fleece is an inverted triangle shape.

[0040] The beneficial effects of the present invention are:

[0041] 1. The inverted triangle tip velvet structure in the present invention amplifies the mechanical vibration of a 0.1mm crack in the pipeline by more than 3 times through the geometric stress concentration effect. The acoustic impedance matching characteristics of the nano-filling layer reduce the signal transmission loss by 40%, and can also preliminarily filter out excess noise signals.

[0042] 2. The present invention forms a distributed sensing array through a high-density winding strategy of more than 30 turns per meter. Combined with optical time-domain reflectometry positioning technology, the spatial positioning accuracy of abnormal points in fire protection pipelines is improved to 5 cm, and the missed detection rate is reduced by 82% compared with the conventional sparse winding method. It is especially suitable for vulnerable parts such as elbows and valves.

[0043] 3. The present invention uses real-time judgment of three-level thresholds, namely normal g<40%, warning 40%≤g≤60%, and alarm g>60%, to make the response delay of the warning design less than 50ms, and supports 10km pipeline online monitoring. It can automatically adjust the threshold parameters according to steel pipes or PVC pipes, greatly reducing false triggering caused by environmental interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a cross-sectional view of the optical fiber of the present invention;

[0045] Figure 2 This is a flow chart of the fire protection pipeline monitoring method based on optical fiber sensing in the present invention;

[0046] Figure 3This is a comparison diagram of the spectrum signals of a normal pipeline, a blocked pipeline, and a defective pipeline in the present invention;

[0047] Figure 4 This is a comparison diagram of the wavelet energy ratios of a normal pipeline, a blocked pipeline, and a defective pipeline in the present invention;

[0048] Figure 5 is a relationship diagram between the abnormal coefficient g and the maximum wavelet energy ratio m of the present invention;

[0049] Figure 6 A graph showing the relationship between the abnormal coefficient g and the number of wavelet packet nodes n of the maximum wavelet energy ratio of the present invention;

[0050] Figure 7 This is a relationship diagram of the abnormal coefficient g, the maximum wavelet energy ratio m, and the number of wavelet packet nodes n of the maximum wavelet energy ratio in the present invention.

[0051] In the figure: 1. Fiber core; 2. Inner cladding; 3. Tip fleece; 4. Nano-filling layer; 5. Outer cladding. DETAILED DESCRIPTION

[0052] The present invention will be described and explained in detail below with reference to the accompanying drawings.

[0053] Example 1

[0054] like Figure 1 As shown, the optical fiber implementing the method for monitoring fire pipelines based on optical fiber sensing comprises a core 1, an inner cladding 2, a tip velvet 3, a nanofilling layer 4, and an outer cladding 5. Several tip velvets 3 are disposed within the nanofilling layer 4, and several of these tip velvets 3 are located on the outer surface of the inner cladding 2. The cross-sectional profile of the tip velvet 3 is an inverted triangle. This inverted triangle tip velvet 3 structure enhances vibration sensitivity through geometric stress concentration. The nanofilling layer 4 reduces acoustic impedance, efficiently transmitting pipeline microvibrations to the core 1. This improves sensitivity by over 30% compared to traditional optical fibers. The inverted triangle cross-section exhibits significant resonance in the 0-250Hz frequency range, enabling detection of pipeline cracks ≤0.1mm.

[0055] The inverted triangle-shaped tip velvet 3 is laser-engraved with micron-level precision, maximizing the geometric stress concentration effect. The nanofilling layer 4, composed of silicon dioxide nanoparticles, is deposited onto the surface of the inner cladding 2 via chemical vapor deposition, forming a uniform and dense nanostructure.

[0056] Furthermore, fiber sensitivity testing was calibrated using a standard vibration source with a frequency range of 1-5 kHz and adjustable amplitude. Crack detection capability testing was conducted at temperatures ranging from 20°C to 80°C and pressures from 0-10 MPa to verify the fiber's stability in extreme environments.

[0057] Compared with traditional optical fibers, the sensitivity of this optical fiber is increased by 35%. At the same time, it can also filter out noise, especially in the 0-250Hz frequency band, where the resonance effect is significant. It can detect pipeline cracks as small as 0.08mm, demonstrating excellent micro-vibration detection capabilities.

[0058] like Figure 2 As shown, the fire protection pipeline monitoring method based on optical fiber sensing is to wind the optical fiber on the fire protection pipeline to be monitored, wherein one end of the optical fiber is connected to the light source output end, and the other end of the optical fiber is connected to the vibration signal acquisition end. The number of turns of the optical fiber wound on the fire protection pipeline is no less than 30 turns per meter of the fire protection pipeline. Signals of different frequency bands are output by the light source output end, and then the vibration signal acquisition end collects the spectrum signal of the fire protection pipeline. The signals of different frequency bands are signals of any frequency band from 0 to 250 Hz. Among them, the high-density winding of ≥30 turns per meter forms a distributed sensing grid, and spatial positioning is achieved through the principle of optical time domain reflection. Each turn of optical fiber is equivalent to an independent sensor, which increases the monitoring resolution to 5 cm level. A single meter of pipeline covers 30 monitoring points, and the missed detection rate is reduced by 82% compared with the sparse winding method.

[0059] like Figure 2-4 As shown, the spectrum signal is normalized to map the amplitude data of the spectrum signal to [0, 1]. The normalization formula is:

[0060] ;

[0061] Where, d i represents the amplitude data after normalization of the spectrum signal; X i Represents the original vibration signal collected in the i-th frequency band.

[0062] The db6 wavelet basis function is used to perform wavelet packet decomposition on the normalized data and calculate the wavelet energy ratio of each wavelet packet node; the db6 wavelet basis function matches the transient impact characteristics of the fire pipe vibration signal, and the wavelet packet decomposition achieves full spectrum coverage in 16 sub-bands from 0 to 250 Hz. The energy ratio calculation highlights the fault characteristic frequency band, so Figure 4 It is proved that two types of failure modes, blockage and defect, can be distinguished. The characteristic of blockage is energy dispersion, and the characteristic of defect is energy peak displacement.

[0063] Among them, the signal is decomposed by wavelet packets and the energy proportion of each frequency band is calculated respectively. Since the total energy of the signal remains unchanged, the energy proportion of different frequency bands is:

[0064] ;in ;

[0065] Where N is the signal length, d i(k) is the value of the kth sampling point corresponding to the i-th frequency band after signal decomposition; E i represents the energy of the i-th frequency band; the total signal energy E is:

[0066] ;

[0067] The wavelet energy proportion m of the corresponding i-th frequency band i for:

[0068] .

[0069] like Figure 3-4 As shown, the original spectrum signal of the normal pipeline is as follows Figure 3 As shown in a, after normalization and wavelet packet decomposition and calculation of the wavelet energy ratio of each wavelet packet node, its wavelet energy is as follows: Figure 4 As shown in a.

[0070] The collected deformation information is transformed by wavelet packet transform using db6 wavelet basis function, and the energy and proportion of each of the 16 wavelet packet nodes are calculated. The results are as follows: Figure 4 As shown in the figure, it can be seen that the overall morphology of the wavelet nodes of pipelines with different defects also shows obvious differences.

[0071] like Figure 4 As shown, the high-energy wavelet nodes for a normal pipeline are concentrated at node 7. When a minor defect occurs inside the pipeline, the energy is concentrated at node 6, clearly distinguishing it from a normal pipeline. For a severely defected pipeline, high energy peaks appear at nodes 6, 8, 9, 13, and 16, and the energy is more dispersed. For a slightly blocked pipeline, the number of high-energy nodes increases, and the energy is not concentrated at a single point but dispersed between nodes 2-6. Similarly, severely blocked pipelines exhibit a strong response within node 10.

[0072] like Figure 2-7 As shown in FIG, a fire monitoring model is generated based on the relationship between the number of wavelet packet nodes of the maximum wavelet energy ratio, the maximum wavelet energy ratio, and the state of the fire pipeline, and the abnormal coefficient of the monitored fire pipeline is calculated through the fire monitoring model; the specific steps of generating the fire monitoring model are:

[0073] S501. Evaluate the abnormality coefficient of the fire protection pipeline according to the degree of blockage and defect of the fire protection pipeline;

[0074] The abnormality coefficient g of the monitored fire protection pipeline can be expressed by the abnormality degree ranking of the sample fire protection pipelines.

[0075] For example, if data from 100 fire protection pipeline samples is collected and, after manual evaluation, the abnormality of a certain fire protection pipeline exceeds the data from the other 50 samples, then the abnormality coefficient g of the monitored fire protection pipeline is 50%.

[0076] Other data are calculated and fitted using data collected by the device itself.

[0077] S502, establishing relational feature 1 based on the blockage degree of the fire protection pipeline and the number of wavelet packet nodes of the maximum wavelet energy ratio;

[0078] A mathematical model is established for the relationship between the abnormal coefficient g and the maximum wavelet energy ratio m, such as Figure 5 As shown, the red dots in the figure are the distribution of the 100 samples collected, so there is formula 1:

[0079] ;

[0080] In the above formula 1, k represents an empirical constant for adjusting the sensitivity of the above model.

[0081] S503, establishing relationship feature 2 based on the blockage degree of the fire protection pipeline and the wavelet energy ratio;

[0082] A mathematical model is established for the relationship between the abnormal coefficient g and the number of wavelet packet nodes n of the maximum wavelet energy ratio, such as Figure 6 As shown in the figure, the red dots are the distribution of the 100 samples collected, so there is formula 2:

[0083] ;

[0084] In the above formula 2, k represents the empirical constant for adjusting the sensitivity of the above model;

[0085] S504: Based on the above relationship feature 1 and relationship feature 2, a fire monitoring model is fitted and generated.

[0086] A mathematical model is established for the relationship between the abnormal coefficient g and the maximum wavelet energy ratio m and the number of wavelet packet nodes n with the maximum wavelet energy ratio, such as Figure 7 As shown, the model shows that the maximum wavelet energy ratio m and the number of wavelet packet nodes n of the maximum wavelet energy ratio are inversely correlated. Combining the characteristic relationship of the above formula 1 and formula 2, we have:

[0087] g=(Formula 1)×(Formula 2);

[0088] The fire monitoring model is generated to calculate the abnormal coefficient of the monitored fire pipeline. The specific formula is:

[0089] ;

[0090] Where g represents the abnormal coefficient of the monitored fire pipeline; n represents the number of wavelet packet nodes with the maximum wavelet energy ratio; and m represents the maximum wavelet energy ratio.

[0091] Among them, n-7 in this formula means that the high-energy wavelet nodes of a normal pipeline are concentrated at node 7. When n-7 is obviously too large or too small, it means that the pipeline is blocked or defective.

[0092] And, as Figure 4 As shown, the eigenvalue of the ordinate is the wavelet energy ratio. The eigenvalue of the slightly blocked pipe is much larger than that of the severely blocked pipe, and the eigenvalue of the slightly defective pipe is also much larger than that of the severely defective pipe.

[0093] Examples of this formula are:

[0094] like Figure 4 As shown in b, the number of wavelet packet nodes with the maximum wavelet energy ratio is n=4, and the maximum wavelet energy ratio is m=0.38. Then we have:

[0095] ;

[0096] Then, from the above calculations, we can know that Figure 4 The abnormality coefficient of the monitored fire protection pipeline in b is g=60.4%.

[0097] like Figure 4 As shown in e, the number of wavelet packet nodes with the maximum wavelet energy ratio is n=6, and the maximum wavelet energy ratio is m=0.68. Then we have:

[0098] ;

[0099] Then, from the above calculations, we can know that Figure 4 The abnormality coefficient of the monitored fire protection pipeline in e is g=42.5%.

[0100] From the above two sets of data, we can know that when g≥42.5%, there will be abnormal phenomena in the monitored fire protection pipeline.

[0101] In the anomaly coefficient model, (n-7) quantifies the energy peak excursion, and m represents the energy concentration. These two variables work together to determine the severity of the fault. This design achieves a model accuracy of >95%, with g = 42.5% indicating a minor blockage and g = 60.4% indicating a major defect. These two defects can be clearly distinguished.

[0102] When g < 40%, the fire protection pipeline is normal; when 40% ≤ g ≤ 60%, it indicates a minor abnormality in the inner wall of the pipeline; and when g > 60%, it indicates a serious fault. This real-time judgment based on three-level thresholds provides an early warning response delay of less than 50ms, supports online monitoring of pipelines up to 10km long, and automatically adjusts threshold parameters based on steel or PVC pipe material, significantly reducing false triggers caused by environmental interference.

[0103] like Figure 2 As shown, different judgment thresholds are established according to the abnormal coefficient and the state of the fire pipeline, and a fire alarm of the corresponding state is issued when the abnormal coefficient of the monitored fire pipeline exceeds the judgment threshold. The judgment thresholds of blocked pipes and inner wall defective pipes are set by comparing the spectrum signals and abnormal coefficients of blocked pipes, inner wall defective pipes and normal pipes. Figure 3-4 The spectrum comparison sets a three-level threshold: g<40% is normal, 40%≤g≤60% is a slight abnormality and a warning, and when g>60% is a serious fault and an alarm is issued. The threshold can be dynamically adjusted according to the pipeline material to reduce the false alarm rate to below 3%. Different thresholds are used for steel pipes and PVC pipes, and the process threshold for steel pipes is 15% higher than that for PVC pipes.

[0104] The judgment threshold is set based on historical data and expert experience, and a dynamic adjustment strategy is employed to automatically adjust the threshold based on changes in pipeline status. Fire alarms include audible and visual alarms, SMS notifications, and app push notifications to ensure timely transmission of alarm information. Compared to a single-level threshold, multi-level threshold settings reduce the false alarm rate to below 3%. With differentiated threshold settings, the false trigger rate is reduced by 50%, significantly improving system stability.

[0105] like Figure 2 As shown, the fire protection pipeline monitoring system based on optical fiber sensing includes:

[0106] The light source output module inputs the light source signal of any frequency band between 0-250Hz into the optical fiber;

[0107] The vibration signal acquisition module collects the spectrum signal output by the light source output module from the other end of the optical fiber;

[0108] The data processing module normalizes the spectrum information and uses wavelet packets to decompose and calculate the wavelet energy ratio of each wavelet packet node;

[0109] The feature analysis module generates a fire monitoring model based on the number of wavelet packet nodes of the maximum wavelet energy ratio, the relationship between the maximum wavelet energy ratio and the state of the fire pipeline, and calculates the abnormal coefficient of the monitored fire pipeline through the fire monitoring model;

[0110] The fire alarm module will sound an alarm when the abnormal coefficient of the monitored fire pipeline exceeds the set threshold.

[0111] The above five modules work together to achieve closed-loop control, with a real-time processing speed of 200 frames per second. The fire pipeline monitoring system based on fiber optic sensing has a response delay of <50ms, supports online monitoring of 10km pipelines, consumes less than 5W, and can be powered by solar energy.

Claims

1. A fire protection pipeline monitoring method based on optical fiber sensing, characterized in that: The following steps are involved: S1. Wind the optical fiber around the fire protection pipeline to be monitored, with one end of the optical fiber connected to the light source output end and the other end connected to the vibration signal acquisition end; S2. The light source output terminal outputs signals of different frequency bands, and the vibration signal acquisition terminal collects the spectrum signal of the fire protection pipeline; S3, performing normalization processing on the spectrum signal to map the amplitude data of the spectrum signal to the range [0, 1]; S4, using the db6 wavelet basis function to perform wavelet packet decomposition on the normalized data and calculate the wavelet energy ratio of each wavelet packet node; S5. Generate a fire monitoring model based on the number of wavelet packet nodes of the maximum wavelet energy ratio, the maximum wavelet energy ratio, and the state relationship of the fire pipeline, and calculate the abnormality coefficient of the monitored fire pipeline through the fire monitoring model; S6. Establish different judgment thresholds based on the abnormal coefficient and the state of the fire protection pipeline, and issue a fire alarm of the corresponding state when the abnormal coefficient of the monitored fire protection pipeline exceeds the judgment threshold; In step S1, the number of turns of the optical fiber wound on the fire protection pipeline is not less than 30 turns per meter of the fire protection pipeline; In step S4, the signal is decomposed by wavelet packets and the energy proportion of each frequency band is calculated. Since the total energy of the signal remains unchanged, the energy proportion of different frequency bands is: ,in ; Where N is the signal length, d i (k) is the value of the kth sampling point corresponding to the i-th frequency band after signal decomposition; E i represents the energy of the i-th frequency band; the total signal energy E is: ； The wavelet energy proportion m of the corresponding i-th frequency band i for: ； In step S5, the specific steps of generating the fire monitoring model are: S501. Evaluate the abnormality coefficient of the fire protection pipeline according to the degree of blockage and defect of the fire protection pipeline; S502, establishing relational feature 1 based on the blockage degree of the fire protection pipeline and the number of wavelet packet nodes of the maximum wavelet energy ratio; S503, establishing relationship feature 2 based on the blockage degree of the fire protection pipeline and the wavelet energy ratio; S504: Fit and generate a fire monitoring model based on the relationship feature 1 and the relationship feature 2; The fire monitoring model is generated to calculate the abnormal coefficient of the monitored fire pipeline. The specific formula is: ； Where g represents the abnormal coefficient of the monitored fire protection pipeline; n represents the number of wavelet packet nodes with the maximum wavelet energy ratio; m represents the maximum wavelet energy ratio; When g<40%, it means that the fire protection pipeline is in normal condition; when 40%≤g≤60%, it means that the inner wall of the fire protection pipeline has a slight abnormality; when g>60%, it means that the inner wall of the fire protection pipeline has a serious fault.

2. The fire protection pipeline monitoring method based on optical fiber sensing according to claim 1 is characterized in that: In step S2, the signals of different frequency bands are signals of any frequency band between 0-250 Hz.

3. The fire protection pipeline monitoring method based on optical fiber sensing according to claim 1 is characterized in that: In step S6, the judgment thresholds for the blocked pipe and the pipe with inner wall defect are set based on the comparison of the spectrum signals and abnormal coefficients of the blocked pipe, the pipe with inner wall defect and the normal pipe.

4. A system for implementing the fire protection pipeline monitoring method based on optical fiber sensing according to any one of claims 1 to 3, characterized in that: include: The light source output module inputs the light source signal of any frequency band between 0-250Hz into the optical fiber; The vibration signal acquisition module collects the spectrum signal output by the light source output module from the other end of the optical fiber; The data processing module normalizes the spectrum information and uses wavelet packets to decompose and calculate the wavelet energy ratio of each wavelet packet node; The feature analysis module generates a fire monitoring model based on the number of wavelet packet nodes of the maximum wavelet energy ratio, the relationship between the maximum wavelet energy ratio and the state of the fire pipeline, and calculates the abnormal coefficient of the monitored fire pipeline through the fire monitoring model; The fire alarm module will sound an alarm when the abnormal coefficient of the monitored fire pipeline exceeds the set threshold.

5. An optical fiber for implementing the fire protection pipeline monitoring method based on optical fiber sensing according to any one of claims 1 to 3, comprising a fiber core (1), an inner cladding (2), a tip fleece (3), a nano-filling layer (4) and an outer cladding (5), characterized in that: A plurality of pointed velvets (3) are provided inside the nano-filling layer (4), and the plurality of pointed velvets (3) are located on the outer surface of the inner cladding (2), and the cross-sectional profile of the pointed velvets (3) is in the shape of an inverted triangle.

Images