Fire-fighting pipeline monitoring method and system based on optical fiber sensing and optical fiber

Abstract

The invention relates to the technical field of optical fiber sensing, and particularly discloses a fire-fighting pipeline monitoring method and system based on optical fiber sensing and an optical fiber. The method comprises the steps that the optical fiber is wound on a fire-fighting pipeline needing to be monitored, one end of the optical fiber is connected with a light source output end, and the other end of the optical fiber is connected with a vibration signal acquisition end; the light source output end outputs signals of different frequency bands, and the vibration signal acquisition end acquires frequency spectrum signals of the fire-fighting pipeline; performing normalization processing on the frequency spectrum signal, and mapping amplitude data of the frequency spectrum signal to be between [0, 1]; and performing wavelet packet decomposition on the normalized data by using a db6 wavelet basis function, and calculating the wavelet energy ratio of each wavelet packet node. According to the inverted triangular tip velvet structure, mechanical vibration of 0.1 mm-level cracks of the pipeline is amplified by more than three times through the geometric stress concentration effect, the signal transmission loss is reduced by 40% through the acoustic impedance matching characteristic of the nanometer filling layer, and redundant noise signals can be preliminarily filtered.

Description

Technical Field

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

[0002] Pipeline transportation has the advantages of large transportation capacity, less land occupation, short construction period, safety and reliability, strong continuity, and high efficiency. After years of development, it has become one of the five major transportation means and plays an irreplaceable role in the development of people's living standards and the national economic construction. As a widely used resource transportation device, a large amount of fresh water, fuel, crude oil, and natural gas are transported through pipelines, and various pipelines have gradually become important infrastructure.

[0003] Among them, a fire pipeline refers to a pipeline material used for fire protection, connecting fire-fighting equipment and appliances, and transporting fire-fighting water, gas, or other media. Due to special requirements, the thickness and material of the fire pipeline have special requirements and are painted red. During the operation of the pipeline, various pipeline anomalies will inevitably occur, including pipeline blockage, pipeline corrosion and defect, and pipeline breakage and leakage.

[0004] Chinese Patent CN118112713A discloses a marine monitoring thin-diameter bending-sensing optical fiber, belonging to the technical field of optical fiber sensors. It includes: a fiber core, an inner cladding, a porous nano-ring, and an outer cladding that are respectively covered around the fiber core from the inside to the outside, and a fiber grating inscribed inside the bending-sensing optical fiber. In the technology disclosed in the above patent, if the sensing optical fiber is directly applied to a fire pipeline as an optical fiber sensor, the signals collected by the optical fiber sensor often contain more noise signals and lack the ability to perform preliminary physical filtering on the noise signals.

[0005] In the prior art, there is a multi-risk early warning system for long-tunnel fire pipelines based on optical fiber sensing. The Brillouin light and Rayleigh light detection unit realizes the combined detection of Brillouin light and Rayleigh light on the optical fiber group based on a dual-wavelength laser, and combines optical and electrical modulation devices to realize the common detection of strain, vibration, and temperature signals; the optical fiber group is tightly attached to the fire pipeline and includes tight optical fibers and loose optical fibers, where the tight optical fibers are used to detect the strain signals at the fire pipeline, and the loose optical fibers are used to detect the temperature and vibration signals at the fire pipeline; the fault determination unit determines the type of fault of the fire pipeline based on the output signals of the optical fiber group. In the above prior art, different judgment thresholds cannot be set according to the types of pipeline anomalies such as pipeline blockage or pipeline defect, and the effect of setting preliminary early warning or alarm according to different blockage degrees or defect degrees cannot be achieved. Summary of the Invention

[0006] In view of the deficiencies in the above prior art, the present invention provides a method and system for monitoring a fire pipeline based on optical fiber sensing, and an optical fiber, so as 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 damage.

[0007] The technical solution adopted by the present invention to solve its technical problems is as follows: The method for monitoring a fire pipeline based on optical fiber sensing includes the following steps: S1. Wind the optical fiber around the fire pipeline to be monitored, where 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; S2. Output signals of different frequency bands from the light source output end, and then collect the spectrum signals of the fire pipeline by the vibration signal acquisition end; S3. Perform normalization processing on the spectrum signals to map the amplitude data of the spectrum signals to between [0, 1]; S4. Use the db6 wavelet basis function to perform wavelet packet decomposition on the data after normalization processing and calculate the wavelet energy ratios of its respective wavelet packet nodes; S5. Generate a fire monitoring model based on the relationship between the number of wavelet packet nodes with the maximum wavelet energy ratio, the maximum wavelet energy ratio, and the state of the fire pipeline, and calculate the anomaly coefficient of the monitored fire pipeline through the fire monitoring model; S6. Establish different judgment thresholds according to the anomaly coefficient and the state of the fire pipeline, and issue a fire alarm for the corresponding state when the anomaly coefficient of the monitored fire pipeline exceeds the judgment threshold.

[0008] Further, in step S1, the number of winding turns of the optical fiber around the fire pipeline is such that the optical fiber wound around each meter of the fire pipeline is not less than 30 turns.

[0009] Further, in step S2, the signals of different frequency bands are signals in any frequency band from 0 to 250 Hz.

[0010] Further, in step S4, perform wavelet packet decomposition on the signal and calculate the energy proportion of each frequency band respectively. Since the total energy of the signal always remains unchanged, the energy proportion in different frequency bands is: ; where ; In the formula, N is the signal length, d i (k) is the value of the kth sampling point corresponding to the ith frequency band after signal decomposition; E i represents the energy of the ith frequency band; the total signal energy E is: ; The corresponding wavelet energy proportion m of the ith frequency bandi is: .

[0011] Furthermore, in the step S5, the specific steps for generating the fire monitoring model are as follows: S501. Evaluate the anomaly coefficient of the fire pipeline according to the blockage degree and defect degree of the fire pipeline; S502. Establish relationship feature 1 according to the number of wavelet packet nodes of the blockage degree of the fire pipeline and the maximum wavelet energy ratio; S503. Establish relationship feature 2 according to the blockage degree of the fire pipeline and the wavelet energy ratio; S504. Generate a fire monitoring model by fitting based on the above relationship feature 1 and relationship feature 2.

[0012] Furthermore, the generated fire monitoring model calculates the anomaly coefficient of the monitored fire pipeline. The specific formula is: ; In the formula, g represents the anomaly coefficient of the monitored fire pipeline; n represents the number of wavelet packet nodes of the maximum wavelet energy ratio; m represents the maximum wavelet energy ratio.

[0013] Furthermore, when g < 40%, it indicates that the state of the fire pipeline is normal; when 40% ≤ g ≤ 60%, it indicates that the inner wall of the fire pipeline is slightly abnormal; when g > 60%, it indicates that the inner wall of the fire pipeline has a serious fault.

[0014] Furthermore, in the step S6, the judgment thresholds of the blocked pipeline and the inner wall defective pipeline are set by comparing the spectral signals and anomaly coefficients of the blocked pipeline, the inner wall defective pipeline and the normal pipeline.

[0015] The fire pipeline monitoring system based on optical fiber sensing includes: A light source output module that inputs a light source signal in any frequency band from 0 to 250 Hz into the optical fiber; A vibration signal acquisition module that acquires the spectral signal output by the light source output module from the other end of the optical fiber; A data processing module that performs normalization processing on the spectral information, uses wavelet packet decomposition, and calculates the wavelet energy ratio of each wavelet packet node; A feature analysis module that generates a fire monitoring model according to 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 calculates the anomaly coefficient of the monitored fire pipeline through the fire monitoring model; A fire alarm module that gives an alarm when the anomaly coefficient of the monitored fire pipeline exceeds the set threshold.

[0016] The optical fiber for implementing the fire pipeline monitoring method based on optical fiber sensing includes a core, an inner cladding, tip fuzz, a nano-filling layer, and an outer cladding. A plurality of tip fuzz are arranged inside the nano-filling layer, and the plurality of tip fuzz are located on the outer surface of the inner cladding. The cross-sectional profile of the tip fuzz is in an inverted triangle shape.

[0017] The beneficial effects of the present invention are as follows: 1. In the present invention, the inverted triangle tip fuzz structure amplifies the mechanical vibration of pipeline cracks at the 0.1 mm level by more than 3 times through the geometric stress concentration effect. The acoustic impedance matching characteristic of the nano-filling layer reduces the signal transmission loss by 40%, and can also preliminarily filter out redundant noise signals.

[0018] 2. The present invention forms a distributed sensing array through a high-density winding strategy of more than 30 turns per meter, and combines the optical time domain reflectometry positioning technology to improve the spatial positioning accuracy of abnormal points on the fire pipeline to 5 cm. The missed detection rate is reduced by 82% compared with the conventional sparse winding method, and it is especially suitable for vulnerable parts such as elbows and valves.

[0019] 3. The present invention makes the response delay of the early warning design <50 ms through real-time judgment of three-level thresholds, that is, normal g < 40%, early warning 40% ≤ g ≤ 60%, and alarm g > 60%, and supports online monitoring of a 10-km pipeline. It can automatically adjust the threshold parameters according to steel pipes or PVC pipes, greatly reducing false triggers caused by environmental interference. Description of the Drawings

[0020] Figure 1 is a cross-sectional view of the optical fiber in the present invention; Figure 2 is a flowchart of the fire pipeline monitoring method based on optical fiber sensing in the present invention; Figure 3 is a comparison diagram of spectral signals of normal pipelines, blocked pipelines, and defective pipelines in the present invention; Figure 4 is a comparison diagram of the proportion of wavelet energy of normal pipelines, blocked pipelines, and defective pipelines in the present invention; Figure 5 is a relationship diagram between the abnormal coefficient g and the maximum wavelet energy ratio m in the present invention; Figure 6 is a relationship diagram between the abnormal coefficient g and the wavelet packet node number n of the maximum wavelet energy ratio in the present invention; Figure 7 is a relationship diagram between the abnormal coefficient g and the maximum wavelet energy ratio m and the wavelet packet node number n of the maximum wavelet energy ratio in the present invention.

[0021] In the figure: 1. Core; 2. Inner cladding; 3. Tip fuzz; 4. Nano-filling layer; 5. Outer cladding. Detailed Embodiments

[0022] The present invention will be specifically described and illustrated below with reference to the accompanying drawings.

[0023] Embodiment 1 As Figure 1 shown, the optical fiber for implementing the fire pipeline monitoring method based on optical fiber sensing includes a fiber core 1, an inner cladding 2, a tip fluff 3, a nano-filling layer 4, and an outer cladding 5. A number of tip fluffs 3 are arranged inside the nano-filling layer 4, and the number of tip fluffs 3 is located on the outer surface of the inner cladding 2. The cross-sectional profile of the tip fluff 3 is in an inverted triangle shape. This inverted triangle tip fluff 3 structure enhances vibration sensitivity through the geometric stress concentration effect. The nano-filling layer 4 reduces the acoustic impedance, enabling the efficient conduction of pipeline micro-vibrations to the fiber core 1, with the sensitivity increased by more than 30% compared to traditional optical fibers. The inverted triangle cross-section has a significant resonance effect in the frequency band of 0 - 250 Hz and can detect pipeline cracks ≤0.1 mm.

[0024] The inverted triangle tip fluff 3 is made by a laser engraving process with an engraving accuracy reaching the micron level to ensure the maximization of the geometric stress concentration effect. The nano-filling layer 4 is composed of silica nanoparticles and is filled on the surface of the inner cladding 2 through a chemical vapor deposition process to form a uniform and dense nano-structure.

[0025] Moreover, the sensitivity test of the optical fiber is calibrated using a standard vibration source, and the vibration frequency range covers 1 - 5 kHz with adjustable amplitude. The crack detection ability test is carried out under different temperatures from 20°C to 80°C and different pressures from 0 to 10 MPa to verify the stability of the optical fiber in extreme environments.

[0026] Compared with traditional optical fibers, the sensitivity of this optical fiber is increased by 35%. At the same time, it can also filter noise. Especially in the frequency band of 0 - 250 Hz, the resonance effect is significant, and it can detect pipeline cracks of 0.08 mm, showing excellent micro-vibration detection ability.

[0027] As Figure 2 shown, for the fire pipeline monitoring method based on optical fiber sensing, the optical fiber is wound around the fire pipeline to be monitored. One end of the optical fiber is connected to the output end of the light source, 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 around the fire pipeline is such that the optical fiber wound per meter of the fire pipeline is not less than 30 turns. Signals of different frequency bands are output from the output end of the light source, and then the spectrum signals of the fire pipeline are collected by the vibration signal acquisition end. The signals of different frequency bands are signals in any frequency band within 0 - 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 optical time domain reflectometry principle. Each turn of the optical fiber is equivalent to an independent sensor, improving the monitoring resolution to the 5 cm level. Each meter of the pipeline covers 30 monitoring points, and the missed detection rate is reduced by 82% compared with the sparse winding method.

[0028] AsFigure 2-4 As shown, the spectral signal is normalized to map the amplitude data of the spectral signal between [0, 1]; among them, the formula for normalization is: ; In the formula, d i represents the amplitude data after normalizing the spectral signal; X i represents the original vibration signal collected in the i-th frequency band.

[0029] And 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 vibration signal of the fire pipeline, and the wavelet packet decomposition achieves full-spectrum coverage in 16 sub-bands of 0 - 250 Hz. The calculation of the energy ratio highlights the fault characteristic frequency band. Therefore Figure 4 it is proved that two types of fault modes, blockage and defect, can be distinguished. The characteristic of blockage is energy dispersion, and the characteristic of defect is energy peak displacement.

[0030] Among them, wavelet packet decomposition is performed on the signal and the energy proportion of each frequency band is calculated separately. Since the total energy of the signal always remains unchanged, the energy proportion in different frequency bands is: ; among them ; In the formula, N is the signal length, and d i (k) is the value of the k-th 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 i corresponding to the i-th frequency band is: .

[0031] As Figure 3-4 shown, the original spectral signal of the normal pipeline is as shown in a of Figure 3 . After normalization, wavelet packet decomposition is performed and the wavelet energy ratio of each wavelet packet node is calculated, and its wavelet energy ratio is as shown in a of Figure 4 .

[0032] And the wavelet packet transform is performed on the collected deformation information through the db6 wavelet basis function, and the energy and proportion of each of the 16 wavelet packet nodes obtained are calculated respectively. The results are as shown in Figure 4 . It can be seen that obvious differences also appear in the overall morphology of the wavelet nodes shown by the pipelines with different defects.

[0033] As Figure 4As shown in the figure, the high-energy wavelet nodes of the normal pipeline are concentrated at node 7. When there is a slight defect inside the pipeline, the energy is concentrated at node 6, showing a relatively obvious distinction from the normal pipeline. For pipelines with severe defects, peaks with relatively high energy ratios appear at nodes 6, 8, 9, 13, and 16, and the energy is relatively dispersed. For slightly blocked pipelines, the number of nodes with high ratios increases, and the energy is not concentrated at a certain point but dispersed between nodes 2 - 6. Similarly, severely blocked pipelines have strong responses within node 10.

[0034] As Figure 2-7 shown, a fire monitoring model is generated based on the number of wavelet packet nodes with the maximum wavelet energy ratio and the maximum wavelet energy ratio, and the relationship with the state of the fire pipeline. The abnormal coefficient of the monitored fire pipeline is calculated through this fire monitoring model. The specific steps for generating the fire monitoring model are as follows: S501. Evaluate the abnormal coefficient of the fire pipeline according to the degree of blockage and defect of the fire pipeline; Among them, the abnormal coefficient g of the monitored fire pipeline can be represented by the ranking of the abnormal degree of the sample fire pipeline.

[0035] For example, if data of 100 samples of fire pipelines are collected, and after manual evaluation, the abnormal degree of a certain fire pipeline exceeds the data in other 50 samples, then it means that the abnormal coefficient g of the monitored fire pipeline = 50%.

[0036] For other data, the data collected by the device itself is used for subsequent calculation and fitting statistics.

[0037] S502. Establish relationship feature 1 according to the degree of blockage of the fire pipeline and the number of wavelet packet nodes with the maximum wavelet energy ratio; Establish a mathematical model for the relationship between the abnormal coefficient g and the maximum wavelet energy ratio m. As Figure 5 shown, the red dots in the figure are the distributions of 100 samples collected. Then there is formula 1: ; In the above formula 1, k represents an empirical constant for adjusting the sensitivity of the above model.

[0038] S503. Establish relationship feature 2 according to the degree of blockage of the fire pipeline and the wavelet energy ratio; Establish a mathematical model for the relationship between the abnormal coefficient g and the number of wavelet packet nodes n with the maximum wavelet energy ratio. As Figure 6 shown, the red dots in the figure are the distributions of 100 samples collected. Then there is formula 2: ; In the above formula 2, k represents an empirical constant for adjusting the sensitivity of the above model; S504. Generate a fire monitoring model by fitting based on the above relationship features 1 and relationship features 2.

[0039] Establish a mathematical model for the relationship between the anomaly coefficient g and the maximum wavelet energy ratio m and the number of wavelet packet nodes n of the maximum wavelet energy ratio. As Figure 7 shown, it can be known through this model 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 relationships of Formula 1 and Formula 2 above, then there is: g = (Formula 1) × (Formula 2); The generated fire monitoring model calculates the anomaly coefficient of the monitored fire pipeline. The specific formula is: ; In the formula, g represents the anomaly coefficient of the monitored fire pipeline; n represents the number of wavelet packet nodes of the maximum wavelet energy ratio; m represents the maximum wavelet energy ratio.

[0040] Among them, n - 7 in this formula means that the high-energy wavelet nodes of the normal pipeline are concentrated on node 7. When n - 7 is significantly larger or smaller, it means that there is a blockage or defect in the pipeline.

[0041] And, as Figure 4 shown, the eigenvalue of the vertical axis is the wavelet energy ratio. The eigenvalue of a slightly blocked pipeline is much larger than that of a severely blocked pipeline, and the eigenvalue of a slightly defective pipeline is also much larger than that of a severely defective pipeline.

[0042] Examples of this formula are: As Figure 4 shown in b, the number of wavelet packet nodes of the maximum wavelet energy ratio is taken as n = 4, and the maximum wavelet energy ratio is taken as m = 0.38. Then there is: ; Then, it can be known from the above calculation that Figure 4 the anomaly coefficient of the monitored fire pipeline in b is g = 60.4%.

[0043] As Figure 4 shown in e, the number of wavelet packet nodes of the maximum wavelet energy ratio is taken as n = 6, and the maximum wavelet energy ratio is taken as m = 0.68. Then there is: ; Then, it can be known from the above calculation that Figure 4 the anomaly coefficient of the monitored fire pipeline in e is g = 42.5%.

[0044] From the above two groups of data, it can be known that when g ≥ 42.5%, there will be an abnormal phenomenon in the monitored fire pipeline.

[0045] In the anomaly coefficient model, (n - 7) quantifies the offset degree of the energy peak, m represents the energy concentration, and the bivariate collaboration determines the severity level of the fault. This design enables the model accuracy to be > 95%. g = 42.5% indicates a slight blockage, and g = 60.4% indicates a severe defect. Slight blockage and severe defect can be clearly distinguished.

[0046] When g < 40%, it indicates that the fire pipeline is in a normal state; when 40% ≤ g ≤ 60%, it indicates a slight anomaly in the inner wall of the fire pipeline; when g > 60%, it indicates a severe fault in the inner wall of the fire pipeline. Through this real-time judgment based on a three-level threshold, the response delay of the early warning is < 50 ms, it supports online monitoring of a 10-km pipeline, and can automatically adjust the threshold parameters according to steel or PVC pipes, greatly reducing false triggers caused by environmental interference.

[0047] As Figure 2 shown, different judgment thresholds are established according to this anomaly coefficient and the state of the fire pipeline, and a fire alarm corresponding to the state is issued when the anomaly coefficient of the monitored fire pipeline exceeds this judgment threshold. By comparing the spectral signals and anomaly coefficients of blocked pipelines, inner wall defective pipelines, and normal pipelines, the judgment thresholds for blocked pipelines and inner wall defective pipelines are set. Based on Figure 3-4 the spectral comparison, three-level thresholds are set: normal when g < 40%, slight anomaly and early warning when 40% ≤ g ≤ 60%, severe fault and alarm when g > 60%, and this threshold can be dynamically adjusted according to the pipeline material, reducing the false alarm rate to below 3%. Differentiated thresholds are used for steel pipes and PVC pipes, and the threshold for steel pipe manufacturing processes is 15% higher than that for PVC pipes.

[0048] The judgment threshold is set according to historical data and expert experience, and a dynamic adjustment strategy is adopted to automatically adjust the threshold according to changes in the pipeline state. The forms of fire alarms include audible and visual alarms, SMS notifications, and APP push notifications to ensure that alarm information is conveyed in a timely manner. Compared with a single-level threshold, setting multi-level thresholds reduces the false alarm rate to below 3%. After setting differentiated thresholds, the false trigger rate is reduced by 50%, significantly improving the system stability.

[0049] As Figure 2 shown, the described fire pipeline monitoring system based on fiber optic sensing includes: A light source output module that inputs a light source signal in any frequency band from 0 - 250 Hz into the optical fiber; A vibration signal acquisition module that collects the spectral signal output by the light source output module from the other end of the optical fiber; A data processing module that normalizes the spectral information, uses wavelet packet decomposition, and calculates the wavelet energy ratio of each wavelet packet node; The feature analysis module generates a fire monitoring model based on the wavelet packet node number of the maximum wavelet energy ratio and the relationship between the maximum wavelet energy ratio and the state of the fire pipeline, and calculates the anomaly coefficient of the monitored fire pipeline through this fire monitoring model; The fire alarm module gives an alarm when the anomaly coefficient of the monitored fire pipeline exceeds the set threshold.

[0050] The above five modules cooperate to achieve closed-loop control, with a real-time processing speed of 200 frames per second. Moreover, the fire pipeline monitoring system based on fiber optic sensing has a response delay < 50 ms, supports online monitoring of 10 km pipelines, and has a power consumption < 5 W and can be powered by solar energy.

Claims

1. A fire pipeline monitoring method based on optical fiber sensing, characterized in that, It includes the following steps: S1. Wind the optical fiber around the fire pipeline to be monitored, where one end of the optical fiber is connected to the output end of the light source, and the other end of the optical fiber is connected to the vibration signal acquisition end; S2. Output signals of different frequency bands from the output end of the light source, and then collect the spectrum signals of the fire pipeline by the vibration signal acquisition end; S3. Perform normalization processing on the spectrum signals to map the amplitude data of the spectrum signals between [0, 1]; S4. Use the db6 wavelet basis function to perform wavelet packet decomposition on the data after normalization processing and calculate the wavelet energy ratios of its respective wavelet packet nodes; S5. Generate a fire monitoring model based on the relationship between the wavelet packet node number of the maximum wavelet energy ratio, the maximum wavelet energy ratio, and the state of the fire pipeline, and calculate the anomaly coefficient of the monitored fire pipeline through this fire monitoring model; S6. Establish different judgment thresholds according to the anomaly coefficient and the state of the fire pipeline, and give a fire alarm for the corresponding state when the anomaly coefficient of the monitored fire pipeline exceeds the judgment threshold.

2. The method for monitoring a fire pipeline based on optical fiber sensing according to claim 1, wherein In the step S1, the number of winding turns of the optical fiber around the fire pipeline is such that the optical fiber wound per meter of the fire pipeline is not less than 30 turns.

3. The method for monitoring a fire pipeline based on optical fiber sensing according to claim 2, wherein, In the step S2, the signals of different frequency bands are signals of any frequency band in 0 - 250 Hz.

4. The method for monitoring a fire pipeline based on optical fiber sensing according to claim 2, wherein In the step S4, perform wavelet packet decomposition on the signals and calculate the energy proportion of each frequency band respectively. Since the total energy of the signals always remains unchanged, the energy proportion in different frequency bands is: , where ; where N is the signal length, d i (k) is the value of the k-th 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 proportion m of the wavelet energy in the corresponding i-th frequency band i is as follows: 。 5. The method for monitoring a fire pipeline based on fiber optic sensing according to claim 2, wherein, In the step S5, the specific steps for generating the fire monitoring model are: S501. Evaluate the anomaly coefficient of the fire pipeline according to the blockage degree and defect degree of the fire pipeline; S502. Establish relationship feature 1 according to the blockage degree of the fire pipeline and the wavelet packet node number of the maximum wavelet energy ratio; S503. Establish relationship feature 2 according to the blockage degree of the fire pipeline and the wavelet energy ratio; S504. Fit and generate a fire monitoring model based on the above relationship feature 1 and relationship feature 2.

6. The method for monitoring a fire pipeline based on optical fiber sensing according to claim 2, wherein The calculation formula for the anomaly coefficient of the monitored fire pipeline by the generated fire monitoring model is specifically: ； In the formula, g represents the anomaly coefficient of the monitored fire pipeline; n represents the wavelet packet node number of the maximum wavelet energy ratio; m represents the maximum wavelet energy ratio.

7. The method for monitoring a fire pipeline based on optical fiber sensing according to claim 6, characterized in that, When g < 40%, it indicates that the state of the fire pipeline is normal; when 40% ≤ g ≤ 60%, it indicates that the inner wall of the fire pipeline is slightly abnormal; when g > 60%, it indicates that the inner wall of the fire pipeline has a serious fault.

8. The method for monitoring a fire pipeline based on optical fiber sensing according to claim 2, wherein, In the step S6, compare the spectrum signals and anomaly coefficients of the blocked pipeline, the inner wall defective pipeline and the normal pipeline to set the judgment thresholds for the blocked pipeline and the inner wall defective pipeline.

9. A fire pipeline monitoring system based on optical fiber sensing according to any one of claims 1-8, characterized in that, It includes: A light source output module that inputs a light source signal of any frequency band in 0 - 250 Hz into the optical fiber; A vibration signal acquisition module that collects the spectrum signals output by the light source output module from the other end of the optical fiber; A data processing module that performs normalization processing on the spectrum information, uses wavelet packet decomposition and calculates the wavelet energy ratios of its respective wavelet packet nodes; The feature analysis module generates a fire monitoring model based on the wavelet packet node number of the maximum wavelet energy ratio and the relationship between the maximum wavelet energy ratio and the state of the fire pipeline, and calculates the anomaly coefficient of the monitored fire pipeline through the fire monitoring model; The fire alarm module gives an alarm when the anomaly coefficient of the monitored fire pipeline exceeds the set threshold.

10. An optical fiber for implementing the optical fiber sensing-based fire pipeline monitoring method according to any one of claims 1-8, comprising a core (1), an inner cladding (2), a tip fluff (3), a nano-filling layer (4), and an outer cladding (5), characterized in that, A number of tip furs (3) are arranged inside the nano-filling layer (4), and the number of tip furs (3) is located on the outer surface of the inner cladding layer (2). The cross-sectional profile of the tip fur (3) is in an inverted triangular shape.

Images