Optical fiber leak detection method, apparatus, medium, and product

Abstract

The embodiment of the application relates to the field of optical fiber sensing, and discloses an optical fiber oil leakage detection method, equipment, medium and product; the method comprises the following steps: arranging a sensing optical fiber along a to-be-detected oil pipe, emitting an optical signal to the sensing optical fiber and receiving a reflection signal of the sensing optical fiber; monitoring the change of the reflection signal in real time, and processing the reflection signal through a multi-modal feature fusion extraction method; the multi-modal feature fusion extraction method comprises time-frequency domain feature extraction, spatial domain feature extraction and multi-source data feature fusion; the time-frequency domain feature extraction comprises extracting spectrum features and impact features of oil film diffusion; the spatial domain feature extraction comprises constructing a concentration gradient field of oil film diffusion and extracting spatial distribution features by using a neural network; the multi-source data feature fusion comprises synchronously collecting multi-source sensing data, constructing a feature cross matrix and extracting dimension reduction features; thereby, the interference of stable mechanical noise such as on-site pump valve start-stop and vehicle passing is effectively filtered out, and precise qualitative detection of extremely weak oil leakage events is realized.

Description

Technical Field

[0001] This application relates to the field of fiber optic sensing technology, and in particular to a fiber optic oil leakage detection method, device, medium, and product. Background Technology

[0002] Currently, most of the technical solutions related to oil leak detection are as follows: one is the pressure monitoring method, which focuses on monitoring changes in fluid inside the pipeline, and the ultrasonic detection method, which uses ultrasonic reflection signals for localization; another is the infrared imaging method, which focuses on detecting large-area temperature anomalies through differences in thermal radiation; in addition, there is fiber optic grating technology that provides point-based high-precision measurement, and distributed vibration sensing technology for long-distance monitoring.

[0003] However, the aforementioned technical solutions have several shortcomings in complex oil leak monitoring scenarios such as the petroleum industry. First, existing pressure monitoring is extremely insensitive to minute leaks and is highly susceptible to interference from pipeline flow fluctuations, while ultrasonic detection equipment is often very expensive and requires frequent calibration. Second, with traditional distributed fiber optic sensing technology, after direct deployment, high-viscosity crude oil easily adheres to the surface of the fiber optic encapsulation layer, forming an acoustic and mechanical buffer layer, which severely attenuates the strain and acoustic signals generated by minute leaks, leading to a decrease in the underlying sensing capability. In addition, the conventional uniform spiral winding deployment method has extremely poor adaptability in stress concentration areas such as pipeline bends, easily producing deformation artifacts or directly causing missed signals. Furthermore, existing fiber optic demodulation and data processing mechanisms often rely on single amplitude differential or fixed threshold alarms, which are highly susceptible to severe interference from background noise such as pump and valve start-up and shutdown, and vehicle passage in complex industrial environments, resulting in a high false alarm rate. At the same time, conventional systems lack a joint analysis mechanism for in-depth feature mining in the time-frequency and spatial domains, and cannot accurately separate the low-frequency spectral characteristics of oil film diffusion from the high-frequency impact characteristics of the leak moment. Finally, most existing detection methods operate independently in a single mode, lacking cross-fusion and dimensionality reduction verification mechanisms for multi-source sensor data, making it difficult to stably and accurately confirm minor oil leak events in highly challenging field environments. Summary of the Invention

[0004] One objective of this application is to provide a fiber optic oil leak detection method, device, medium, and product, which at least solves the problems in the prior art where the underlying signal perception is weak due to crude oil adhesion buffering and elbow stress interference, and the feature judgment mechanism has a high false alarm rate under complex environmental interference.

[0005] To achieve the above objectives, some embodiments of this application provide the following aspects:

[0006] In a first aspect, some embodiments of this application provide a method for detecting oil leakage in optical fibers, the method comprising:

[0007] A sensing optical fiber is laid along the oil pipe to be tested, and an optical signal is transmitted into the sensing optical fiber and the reflected signal of the sensing optical fiber is received.

[0008] The changes in the reflected signal are monitored in real time, and the reflected signal is processed by a multimodal feature fusion extraction method;

[0009] The multimodal feature fusion extraction method includes time-frequency domain feature extraction, spatial domain feature extraction, and multi-source data feature fusion;

[0010] The time-frequency domain feature extraction includes extracting the spectral features and impact features of oil film diffusion;

[0011] The spatial domain feature extraction includes constructing the concentration gradient field of the oil film diffusion and extracting spatial distribution features using a neural network;

[0012] The multi-source data feature fusion includes synchronously collecting multi-source sensor data and constructing a feature cross matrix to extract dimensionality-reduced features.

[0013] Secondly, some embodiments of this application also provide an electronic device, the electronic device comprising: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method described above.

[0014] Thirdly, some embodiments of this application also provide a computer-readable medium having computer program instructions stored thereon, which can be executed by a processor to implement the method described above.

[0015] Fourthly, some embodiments of this application also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement the method described above.

[0016] Compared with related technologies, the solution provided in this application introduces a multimodal feature fusion extraction method at the signal processing level, thereby solving the problem of high false alarm rate of single feature judgment mechanism in complex industrial environments. Through time-frequency domain feature extraction, the single dimension of the reflection signal is analyzed in a higher dimension. Short-time Fourier transform is used to extract the spectral features of oil film diffusion, and wavelet transform is combined to capture the impact features of the instantaneous leakage, constructing a combined long- and short-time analysis strategy. This feature extraction mechanism can separate the low-frequency spectral features representing the slow diffusion of the oil film from the high-frequency impact features representing pipe wall rupture or fluid jetting, thereby effectively filtering out interference from steady-state mechanical noise such as pump and valve start-up and shutdown, and vehicle passage, achieving accurate qualitative detection of extremely weak oil leak events.

[0017] Furthermore, a rigorous multi-dimensional cross-validation system is constructed through spatial domain feature extraction and multi-source data feature fusion mechanisms. A neural network is used to extract the spatial distribution features of the concentration gradient field, and simultaneously, multi-source sensor data from sonar, radar, and optics are fused for dimensionality reduction of the feature cross-matrix, overcoming the limitations of traditional single-point or single-modal monitoring. Relying on the mutual verification of multi-dimensional heterogeneous data, the method provided in this application can effectively eliminate false alarm signals from single sensors, significantly reducing the false alarm rate of oil leak detection and providing a high-confidence, high-robust safety monitoring solution for oil transportation and storage networks.

[0018] Furthermore, regarding the underlying physical sensing structure, by coating the surface of the sensing fiber with a superhydrophobic coating having a contact angle greater than 150°, and employing an "8"-shaped arrangement at the bends of the oil pipe under test, the problems of signal attenuation caused by crude oil adhesion and signal distortion in stress concentration areas are effectively solved. This physical layer protection and targeted arrangement structure prevents high-viscosity crude oil from forming an acoustic and mechanical buffer layer on the fiber surface, thereby avoiding signal artifacts caused by localized stress concentration. This arrangement and encapsulation method ensures that minute leakage strain signals can be transmitted to the sensing fiber without loss, fundamentally improving the sensitivity and fidelity of the underlying physical signal acquisition. Attached Figure Description

[0019] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0020] Figure 1 An exemplary flowchart of an optical fiber oil leakage detection method provided in some embodiments of this application;

[0021] Figure 2 An exemplary structural diagram of the electronic device provided for some embodiments of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] Figure 1 An exemplary flowchart of a fiber optic oil leakage detection method provided in some embodiments of this application, the method comprising:

[0024] S101. Lay out a sensing optical fiber along the oil pipe to be tested, transmit an optical signal to the sensing optical fiber, and receive the reflected signal from the sensing optical fiber.

[0025] Specifically, this method is based on a fiber optic sensing system, including a light source module, a sensing fiber, a signal demodulation module, and a data processing unit. The light source module and the sensing fiber are connected via a fiber optic coupler. The sensing fiber is deployed along the oil pipeline or oil storage facility. The signal demodulation module receives the reflected signal from the sensing fiber and transmits it to the data processing unit. The light source module can be a narrow-linewidth laser with a wavelength of 1550 nm and an output power of 10 mW. The sensing fiber can be a single-mode fiber with a polyimide coating and a temperature range of -40℃ to 120℃. The signal demodulation module can be based on a phase-sensitive optical time-domain reflectometer (Φ-OTDR) with a spatial resolution of 1 m and a dynamic range of 60 dB.

[0026] S102. Monitor the changes in the reflected signal in real time, and process the reflected signal using a multimodal feature fusion extraction method.

[0027] Specifically, when an oil leak occurs, the oil comes into contact with the sensing optical fiber, causing a change in the local refractive index, which in turn causes a phase shift in the backscattered Rayleigh light. By receiving and detecting the phase change of the reflected signal through the signal demodulation module, a preliminary perception of the pipeline's physical condition can be achieved, and the leak point can be located.

[0028] S103, The multimodal feature fusion extraction method includes time-frequency domain feature extraction, spatial domain feature extraction, and multi-source data feature fusion.

[0029] Specifically, since single signals are highly susceptible to interference in complex industrial environments, this embodiment performs in-depth analysis of the reflected signal through joint analysis of the above three dimensions. The multimodal feature fusion extraction method includes time-frequency domain feature extraction, spatial domain feature extraction, and multi-source data feature fusion.

[0030] S104. The time-frequency domain feature extraction includes extracting the spectral features and impact features of oil film diffusion.

[0031] Specifically, since oil leakage is a dynamic process, its initial rupture and subsequent diffusion exhibit distinctly different physical characteristics in the time and frequency domains. Therefore, short-time Fourier transform (SFT) is used to extract the spectral characteristics of the oil film diffusion. For example, when crude oil slowly seeps out and diffuses in the surrounding medium, it generates low-frequency fluctuations in the range of 0.1 Hz to 5 Hz; SFT can accurately isolate and quantify these low-frequency spectral characteristics. Simultaneously, wavelet transform is combined to capture the impact characteristics at the moment of leakage. For example, the instant of pipe wall rupture or high-pressure oil injection generates high-frequency transient shock waves; the abrupt peak values ​​of the Daubechies wavelet coefficients can sensitively extract these impact characteristics. Through the combination of SFT and wavelet transform, comprehensive coverage and extraction of steady-state low-frequency diffusion signals and transient high-frequency impact signals are achieved.

[0032] S105. The spatial domain feature extraction includes constructing the concentration gradient field of the oil film diffusion and extracting spatial distribution features using a neural network.

[0033] Specifically, oil leaks not only cause distortions in time-series signals but also create a gradually spreading diffusion field in physical space. A two-dimensional concentration gradient field of the oil film diffusion is constructed by scanning the monitored area using laser-induced fluorescence technology or infrared thermal imaging equipment. After acquiring this two-dimensional concentration gradient field, it is input into a pre-trained convolutional neural network for feature mining. The convolutional neural network can automatically identify and extract deep-level spatial distribution features, such as the edge diffusion rate and area growth rate of the oil film. Based on the extracted spatial distribution features, the authenticity of the leak event can be further confirmed from the perspective of spatial geometric morphology evolution.

[0034] S106. The multi-source data feature fusion includes synchronously collecting multi-source sensor data and constructing a feature cross matrix to extract dimensionality-reduced features.

[0035] Specifically, to eliminate the blind spots and random errors of a single sensing mechanism, a cross-validation mechanism using multi-dimensional physical fields is introduced. Multi-source sensing data from multiple heterogeneous sensors can be collected simultaneously, including: leakage acoustic signature data collected using sonar equipment, oil film surface roughness data detected using radar equipment, and oil film spectral absorption peak data monitored using optical sensors. After acquiring the above multi-source sensing data, the time-frequency domain features, spatial domain features, and the above multi-source sensing data are time-stamp aligned and dimension-mapped to construct a feature cross-matrix. Subsequently, the feature cross-matrix is ​​dimensionality-reduced to remove redundant noise and weakly correlated variables, extracting the dimensionality-reduced features that best characterize the essence of the oil leak, thereby completing the deep fusion of multi-dimensional physical features.

[0036] In this embodiment, a multimodal feature fusion extraction method is introduced to extend the analysis of a single reflection signal to the joint analysis of time-frequency domain, spatial domain, and multi-source sensor data. Time-frequency domain feature extraction can accurately separate low-frequency leakage and high-frequency rupture features, effectively filtering out steady-state mechanical noise; spatial domain feature extraction and multi-source data feature fusion construct a rigorous cross-validation mechanism, using the mutual verification of multidimensional heterogeneous data to eliminate false alarms. The above scheme breaks through the limitations of traditional single-modal monitoring in complex industrial environments, significantly reduces the false alarm rate of oil leak detection, and achieves high-sensitivity and high-confidence monitoring of minor leakage events.

[0037] In one embodiment, the time-frequency domain feature extraction includes extracting the spectral features using short-time Fourier transform and extracting the impulse features using wavelet transform.

[0038] Specifically, considering the slowly varying physical characteristics of oil seepage and oil film diffusion, a short-time Fourier transform (SFT) is used to process the received reflected signal. In the calculation of the SFT, a Hanning window can be used as the sliding window function, and the overlap rate of adjacent windows is set to 50% to 75% to effectively suppress spectral leakage and ensure frequency resolution in the extremely low frequency band. Through the SFT, low-frequency fluctuation data in the 0.1 Hz to 5 Hz range are extracted as the spectral features.

[0039] Simultaneously, considering the non-stationary transient physical characteristics generated during pipe wall rupture or high-pressure fluid ejection, wavelet transform is used to capture the impact characteristics at the moment of leakage. Specifically, the Daubechies 4 (db4) wavelet can be used as the wavelet basis function to perform multi-scale discrete wavelet decomposition of the reflected signal at 4 to 5 levels, separating out the detail coefficients containing high-frequency transient energy, and extracting the abrupt peak of the detail coefficients as the impact characteristics. Wavelet transform has multi-resolution analysis characteristics and can provide extremely high time resolution in the high-frequency band, thereby accurately locking onto weak transient high-frequency signals.

[0040] In this embodiment, by combining short-time Fourier transform (SFT) and wavelet transform, simultaneous and accurate analysis of continuously varying signals and transient abrupt signals is achieved. SFT overcomes the limitation of pure time-domain analysis in capturing the dynamic frequency distribution, accurately pinpointing the unique low-frequency characteristics of oil film diffusion. Wavelet transform compensates for the insufficient time positioning accuracy of traditional Fourier transform when processing non-stationary abrupt signals, keenly capturing the extremely high-frequency impact characteristics of the initial stage of leakage. This analytical strategy completely separates the heterogeneous characteristic signals generated by actual oil leaks from the steady-state mechanical noise commonly found in industrial settings, significantly improving the completeness of feature extraction and the accuracy of oil leak event detection from the underlying signal processing stage.

[0041] In one embodiment, the spatial domain feature extraction includes constructing the concentration gradient field by laser-induced fluorescence or infrared thermal imaging, and extracting the spatial distribution features using a convolutional neural network.

[0042] Specifically, a laser-induced fluorescence device is used to emit excitation light of a specific wavelength into the monitoring area around the oil pipe, exciting molecules such as polycyclic aromatic hydrocarbons in the leaking oil to generate fluorescence signals in a specific frequency band; or an infrared thermal imaging device is used to capture the objective thermal radiation difference in heat capacity between the small leaking oil and the surrounding background medium. After collecting the above fluorescence or thermal radiation signals, pixel-level mapping and intensity calibration are performed to construct the concentration gradient field that can intuitively reflect the oil film distribution state. After obtaining the concentration gradient field, it is input into a pre-trained convolutional neural network for feature mining. The convolutional neural network adopts a deep architecture with residual connections (such as the ResNet-50 model), and through multiple sets of alternately stacked convolutional and pooling layers, it extracts the hidden local edge textures and global geometric evolution rules in the concentration gradient field step by step, and finally outputs the quantified spatial distribution features, such as the edge diffusion rate and area growth rate of the oil film, through a fully connected layer.

[0043] In the above embodiments, a concentration gradient field is constructed using laser-induced fluorescence or infrared thermal imaging technology, compensating for the blind spots of one-dimensional fiber optic sensing in macroscopic physical morphology perception. This expands the single linear monitoring into a three-dimensional planar mapping, achieving complete capture of the diffusion trajectory of leaked substances. A convolutional neural network is introduced to extract spatial distribution features. Utilizing the translation invariance and strong nonlinear fitting capabilities of deep learning models in image pattern recognition, visual interference caused by sudden changes in light, cluttered ground backgrounds, or partial equipment obstruction in industrial settings is automatically filtered out. This processing transforms the complex and difficult-to-quantify physical diffusion phenomenon into extremely precise geometric parameters, providing reliable spatial morphological support for the system.

[0044] In one embodiment, the multi-source sensing data includes sonar data, radar data, and optical data.

[0045] Specifically, to achieve cross-validation of multi-dimensional physical fields, sonar acquisition equipment, radar scanning equipment, and optical monitoring equipment are simultaneously activated. The sonar acquisition equipment acquires acoustic propagation signals in the medium, and extracts underwater leakage acoustic signatures from these signals as sonar data. The radar scanning equipment emits microwave or millimeter-wave electromagnetic waves towards the oil film-covered area and receives the echoes, calculating the phase and amplitude changes of the echoes and extracting spatial scattering characteristics reflecting the surface roughness of the oil film as radar data. The optical monitoring equipment emits a beam of light in a specific band towards the area under test, detects absorption characteristics in the reflection spectrum, and extracts spectral absorption peak data characterizing oil molecule concentration as optical data. After acquiring the aforementioned heterogeneous multi-source sensing data, the extracted spectral features, impact features, and spatial distribution features are time-stamp aligned and spatially mapped with the sonar data, radar data, and optical data to construct a high-dimensional feature cross-matrix. Subsequently, algorithms such as principal component analysis or linear discriminant analysis are used to process the feature cross matrix, remove redundant noise and weakly correlated variables in the feature cross matrix, and extract the core principal components with the largest variance contribution rate as dimensionality reduction features.

[0046] In the above embodiments, by introducing sonar data, radar data, and optical data, simultaneous acquisition and multi-source complementarity of acoustic, electromagnetic, and optical multi-physics information are achieved. Constructing a feature cross-matrix and extracting dimensionality-reduced features not only eliminates the perception blind spots of a single sensor in specific extreme environments but also effectively filters out redundant interference information between different modal data. This multi-source data feature fusion mechanism ensures that the system can make judgments based on rigorous verification of multi-dimensional heterogeneous data, significantly improving the robustness and accuracy of oil leak detection in complex industrial environments.

[0047] In one embodiment, the deployment of the sensing optical fiber along the oil pipe to be measured includes:

[0048] The sensing optical fiber is evenly spirally wound around the straight section of the oil pipe to be tested, and the sensing optical fiber is arranged in a figure-eight shape at the bend of the oil pipe to be tested.

[0049] Specifically, for the straight section of the oil pipe under test, considering the relatively uniform stress distribution of the pipe wall, the sensing optical fiber is laid using a uniform spiral winding method. During the winding operation, a set winding pitch and constant pre-tension stress are maintained, and fixing tape or adhesive is used to ensure that the sensing optical fiber is tightly attached to the outer wall of the oil pipe under test, thereby achieving uniform sensing of minute deformations of the pipe wall in the straight section. For the bends of the oil pipe under test, since fluid turning impacts can easily cause local stress concentration and severe deformation, a figure-eight shaped intersecting path is used for laying. Specifically, the sensing optical fiber is guided to alternately cross the inner pressure surface and the outer tension surface at the bend, forming a symmetrical overlapping physical structure, and the intersection nodes are positioned and solidified.

[0050] In the above embodiments, by implementing differentiated layout schemes for straight pipe sections and bends, the fidelity of the underlying physical signal acquisition is significantly improved. The uniform spiral winding of the straight pipe sections ensures the continuity of monitoring coverage and the consistency of sensing sensitivity along the line; the figure-eight layout at the bends utilizes the symmetrical intersection of physical structures to effectively neutralize the local stress concentration effects caused by fluid impact and pipe bends, preventing baseline drift and signal artifacts caused by uneven structural stress. This layout method ensures that weak oil leakage strain signals can be transmitted losslessly and stably, eliminating interference from specific spatial structures at the physical source and reducing the risk of false alarms in the system.

[0051] In one embodiment, the surface of the encapsulation layer of the sensing optical fiber is coated with a superhydrophobic coating; the contact angle of the superhydrophobic coating is greater than 150°.

[0052] Specifically, considering the generally high viscosity of fluids such as crude oil, a nanoscale surface treatment is performed on the outermost polyimide encapsulation layer of the sensing optical fiber. The specific operation involves preparing a coating solution using a composite of fluoropolymer and nanoparticles, uniformly adhering the coating solution to the encapsulation layer surface via dip coating or continuous spraying, and then curing and cross-linking to form the superhydrophobic coating. By controlling the physical morphology and surface free energy of the micro-nano rough structure in the superhydrophobic coating, the contact angle of the superhydrophobic coating with oil or water molecules is made greater than 150°. When a pipeline leaks and high-viscosity oil comes into contact with the sensing optical fiber, fluid droplets with a contact angle greater than 150° cannot effectively wet and adhere to the surface of the sensing optical fiber, and thus slide off along the direction of force under slight disturbance.

[0053] In the above embodiments, by coating the outermost layer of the sensing fiber with a superhydrophobic coating having a contact angle greater than 150°, the long-term physical adsorption and accumulation of high-viscosity oil on the fiber surface is blocked from the perspective of material physicochemical properties. This anti-adhesion mechanism effectively prevents leaked crude oil from accumulating and solidifying around the sensor, forming an acoustic and mechanical buffer layer, and preventing the absorption and attenuation of weak leakage strain signals and high-frequency rupture shock waves by the local oil contamination buffer layer. The physical encapsulation structure can maintain the cleanliness of the outer wall of the sensing fiber and its interface conduction characteristics for a long time, ensuring that the sensing fiber maintains extremely high low-level sensing sensitivity and signal reception fidelity to extremely weak physical field distortions around the pipeline.

[0054] In one embodiment, the method further includes establishing a virtual model of the sensing fiber deployment structure using digital twin technology, and determining the deployment parameters of the sensing fiber through finite element analysis.

[0055] Specifically, firstly, the three-dimensional geometric data, material mechanical properties, and on-site geological environment data of the oil pipe to be tested are collected. Based on the above data, a high-fidelity virtual model of the oil pipe to be tested and its surrounding sensing fiber optic network structure is generated in a virtual digital space. After establishing the virtual model, the virtual model is meshed, and the finite element method is introduced for multiphysics coupling solution. During the finite element simulation, simulated working condition boundary conditions are applied to the virtual model. These working condition boundary conditions include setting oil leakage pressure under different orifice diameters, local deformation caused by fluid impact, and temperature gradient changes in the external environment. Through simulation calculations, the dynamic process and attenuation law of physical quantities such as stress, strain, and temperature transmitted from the oil pipe wall to the sensing fiber optic network are observed and obtained, thereby accurately capturing the stress and deformation characteristics of the sensing fiber optic network under different working conditions. Based on the simulation feedback results of the finite element analysis, the spatial arrangement variables in the virtual model are continuously iterated and adjusted until the layout parameters that achieve the optimal signal-to-noise ratio of the fiber optic sensing signal are extracted. The specific deployment parameters include: the spiral winding pitch of the sensing optical fiber in the straight pipe section, the figure-eight crossing angle at the bend, the pre-tension strain threshold applied when laying the optical fiber, and the thickness parameters of the encapsulation coating.

[0056] In this embodiment, a virtual model mapping to a physical entity is constructed by combining digital twin technology and finite element analysis, enabling quantitative simulation and deduction of the underlying fiber optic deployment scheme. This processing method can calculate the dynamic process of physical quantity transmission and attenuation under complex operating conditions, effectively avoiding blind trial and error in the physical field and completely overcoming the shortcomings of traditional schemes that rely on manual experience and incur high rework costs. Iterative optimization of spatial deployment parameters based on finite element analysis ensures that the actually deployed sensor network maintains the highest sensing sensitivity to extremely small oil film diffusion and impact characteristics under various harsh operating conditions, significantly improving the robustness of the entire monitoring system in complex industrial environments.

[0057] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this application. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of the algorithm and process, are also within the scope of protection of this application.

[0058] Furthermore, some embodiments of this application also provide an electronic device. The electronic device can be various forms of digital computer, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, etc. The electronic device can also be various forms of mobile devices, such as cellular phones, smartphones, wearable devices, and other similar computing devices.

[0059] The electronic device includes: one or more processors; and a memory storing computer program instructions that, when executed, cause the processor to perform the steps of the methods provided in any one or more of the above embodiments. Figure 2 An exemplary structural diagram of the electronic device is disclosed. The electronic device includes one or more processors 1101, a memory 1102, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple electronic devices can be connected, each providing some of the necessary operations. The components, their connections and relationships, and their functions shown herein are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0060] The electronic device may further include an input device 1103 and an output device 1104. The processor 1101, memory 1102, input device 1103 and output device 1104 may be connected by a bus or other means, as shown in the figure, which is connected by a bus.

[0061] Input device 1103 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the electronic device, such as a touch screen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 1104 may include a display device, auxiliary lighting device (e.g., LED), and haptic feedback device (e.g., vibration motor). The display device may include, but is not limited to, a liquid crystal display, a light-emitting diode display, and a plasma display. In some embodiments, the display device may be a touch screen.

[0062] To provide interaction with the user, the electronic device can be a computer. The computer has: a display device (e.g., a cathode ray tube or LCD monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback); and input from the user can be received in any form (e.g., voice input or tactile input).

[0063] In this embodiment, a computer-readable medium stores a computer program / instructions that, when executed by a processor, implement the steps of the methods provided in any one or more of the above embodiments. This computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into that device. The aforementioned computer-readable medium carries one or more computer-readable instructions.

[0064] The memory 1102 can serve as a non-transitory computer-readable storage medium, used to store non-transitory software programs, non-transitory computer-executable programs, and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 1102, thereby implementing the program instructions / modules corresponding to the methods provided in any one or more of the embodiments described above in this application.

[0065] The memory 1102 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device. Furthermore, the memory 1102 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 1102 may optionally include memory remotely located relative to the processor 1101, and these remote memories can be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0066] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0067] Computer-readable media include permanent and non-permanent, removable and non-removable media, which can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory, static random access memory, dynamic random access memory, other types of random access memory, read-only memory, electrically erasable programmable read-only memory, flash memory or other memory technologies, read-only optical discs, digital versatile optical discs or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0068] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0069] In the above embodiments, all or part of the implementation can be achieved through software, hardware, firmware, or any combination thereof. For example, it can be implemented using an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device. In some embodiments, the software program of this application can be executed by a processor to implement the above steps or functions. Similarly, the software program of this application (including related data structures) can be stored in a computer-readable recording medium, such as RAM memory, magnetic or optical drives, floppy disks, and similar devices. In addition, some steps or functions of this application can be implemented in hardware, for example, as circuitry that cooperates with a processor to perform the various steps or functions.

[0070] The computer program product provided in this application includes one or more computer programs / instructions. When executed by a processor, these computer programs / instructions generate, in whole or in part, the processes or functions described in this application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0071] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-specific system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0072] The scope of this application is defined by the appended claims rather than the foregoing description, and is therefore intended to encompass all variations falling within the meaning and scope of equivalents of the claims. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a device claim may also be implemented by a single unit or device in software or hardware. Terms such as "first," "second," etc., are used only for distinguishing descriptions and do not indicate any particular order, nor should they be construed as indicating or implying relative importance.

[0073] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily made by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.

Claims

1. A method for detecting oil leakage in optical fibers, characterized in that, The method includes: A sensing fiber is laid along the oil pipe to be tested, and an optical signal is transmitted into the sensing fiber and the reflected signal of the sensing fiber is received. The changes in the reflected signal are monitored in real time, and the reflected signal is processed by a multimodal feature fusion extraction method; The multimodal feature fusion extraction method includes time-frequency domain feature extraction, spatial domain feature extraction, and multi-source data feature fusion; The time-frequency domain feature extraction includes extracting the spectral features and impact features of oil film diffusion; The spatial domain feature extraction includes constructing the concentration gradient field of the oil film diffusion and extracting spatial distribution features using a neural network; The multi-source data feature fusion includes synchronously collecting multi-source sensor data and constructing a feature cross matrix to extract dimensionality-reduced features. The process of laying the sensing fiber along the oil pipe to be tested includes: uniformly spirally winding the sensing fiber around the straight section of the oil pipe to be tested, and laying the sensing fiber in a figure-eight shape at the bend of the oil pipe to be tested. The surface of the encapsulation layer of the sensing optical fiber is coated with a superhydrophobic coating; the contact angle of the superhydrophobic coating is greater than 150°.

2. The fiber optic oil leakage detection method according to claim 1, characterized in that, The time-frequency domain feature extraction includes extracting the spectral features using short-time Fourier transform and extracting the impulse features using wavelet transform.

3. The fiber optic oil leakage detection method according to claim 1, characterized in that, The spatial domain feature extraction includes constructing the concentration gradient field through laser-induced fluorescence or infrared thermal imaging, and extracting the spatial distribution features using a convolutional neural network.

4. The fiber optic oil leakage detection method according to claim 1, characterized in that, The multi-source sensor data includes sonar data, radar data, and optical data.

5. The fiber optic oil leakage detection method according to claim 1, characterized in that, The method also includes establishing a virtual model of the sensing fiber optic deployment structure using digital twin technology, and determining the deployment parameters of the sensing fiber optic cable through finite element analysis.

6. An electronic device, characterized in that, The electronic device includes: One or more processors; and A memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method as described in any one of claims 1 to 5.

7. A computer-readable medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 5.

8. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1 to 5.

Images