Optical fiber sensing signal processing method and device and storage medium

Abstract

The invention discloses an optical fiber sensing signal processing method and device and a storage medium. The method comprises the following steps: acquiring a plurality of optical fiber sensing signals; extracting a background signal and an intrusion signal of each optical fiber sensing signal; inputting any one background signal and any one intrusion signal into a preset model for combination and reconstruction to generate a corresponding target optical fiber sensing signal; and inputting each target optical fiber sensing signal into the optical fiber sensing basic large model so as to output a signal feature corresponding to each target optical fiber sensing signal through the optical fiber sensing basic large model. According to the method, useful signals and noise can be accurately separated, the signal identification precision is improved, the identification and extraction capability of optical fiber sensing signals can be enhanced, the interference and influence of environmental noise are reduced, and the generalization capability and accuracy of the model are improved.

Description

Technical Field

[0001] The present application relates to the technical field of optical fiber sensor signal processing, and in particular to an optical fiber sensor signal processing method, device and storage medium. Background Art

[0002] Currently, fiber optic sensing technology has been widely used in many fields, including communications, medical treatment, environmental monitoring, and safety protection. Fiber optic sensors have the advantages of high sensitivity, anti-electromagnetic interference, and long-distance transmission. However, with the complexity of application scenarios and the increase in data volume, traditional fiber optic sensing technology faces many problems in data processing and analysis. The amount of data collected by fiber optic sensors is huge, and the data processing and analysis process is complex, requiring efficient algorithms and powerful computing power. Since fiber optic sensing signals are easily affected by environmental noise and interference, it becomes difficult to accurately identify and extract useful signal features. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a method, device and storage medium for processing optical fiber sensor signals, so as to solve the technical problem that the data processing process of optical fiber sensor in the prior art is complex and susceptible to interference.

[0004] In order to achieve the above-mentioned object, the first aspect of the present application provides a method for processing an optical fiber sensing signal, comprising:

[0005] Acquire multiple optical fiber sensing signals;

[0006] Extracting background signal and intrusion signal of each optical fiber sensing signal;

[0007] Input any background signal and any intrusion signal into the preset model for combined reconstruction to generate the corresponding target optical fiber sensing signal;

[0008] Each target optical fiber sensing signal is input into the optical fiber sensing basic large model, so as to output the signal characteristics corresponding to each target optical fiber sensing signal through the optical fiber sensing basic large model.

[0009] In an embodiment of the present application, the preset model is a diffusion model, and the method also includes a step of training the diffusion model: obtaining historical fiber optic sensor signals of historical intrusion events and influencing factors corresponding to the fiber optic sensing; performing noise processing on the historical fiber optic sensor signals according to the influencing factors to obtain corresponding noisy fiber optic sensor signals; performing denoising processing on the noisy fiber optic sensor signals to obtain corresponding denoised fiber optic sensor signals; and training the diffusion model according to the historical fiber optic sensor signals and the denoised fiber optic sensor signals to obtain a trained diffusion model.

[0010] In an embodiment of the present application, a noisy fiber optic sensor signal is denoised to obtain a corresponding denoised fiber optic sensor signal; a main decoupling factor of the noisy fiber optic sensor signal is determined based on a principal component analysis method, the main decoupling factor including background information, intrusion information and noise; and the noisy fiber optic sensor signal is denoised according to the main decoupling factor to obtain a corresponding denoised fiber optic sensor signal.

[0011] In an embodiment of the present application, the influencing factors include at least one of meteorological data, optical fiber laying method, network hanging method, and wind force level.

[0012] In the embodiment of the present application, the functional expression of the objective function of the diffusion model training process is shown in the following formulas (1) and (2):

[0013]

[0014] Among them, t1 refers to the time step, x0 refers to the real sample drawn from the distribution p(x0) of the training data set, and x t It refers to the sample after adding noise to the real sample at time step t1, ∈ θ refers to the preset model, ∈ θ (x t1 ,t1) refers to the predicted output of the preset model at time step t1, ∈ refers to the added noise, Refers to the preset weighting coefficient.

[0015] In an embodiment of the present application, any background signal and any intrusion signal are input into a preset model for combined reconstruction to generate a corresponding target optical fiber sensing signal, including the target optical fiber sensing signal being calculated according to the following formula (3):

[0016]

[0017] Among them, t2 refers to the time step, x t2-1 It refers to the target optical fiber sensor signal when the time step is t2-1, ∈ θ refers to the preset model, x t2 It refers to the sample after adding noise to the real sample at time step t2, x B is the background signal, x I is the intrusion signal, σ t2 is the noise scale coefficient at time step t2, z is the random variable, a t2 Refers to the preset weighting coefficient.

[0018] In an embodiment of the present application, extracting the background signal and the intrusion signal of each optical fiber sensing signal includes: determining the statistical characteristics of each optical fiber sensing signal; and extracting the background signal and the intrusion signal of each optical fiber sensing signal according to the statistical characteristics.

[0019] In the embodiment of the present application, the statistical feature includes any one of variance, frequency domain characteristics, local vibration characteristics, and significant area.

[0020] A second aspect of the present application provides a device for processing optical fiber sensing signals, comprising:

[0021] a memory configured to store instructions;

[0022] The processor is configured to call instructions from the memory and implement the above-mentioned method for processing optical fiber sensing signals when executing the instructions.

[0023] A third aspect of the present application provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the processing according to the above-mentioned optical fiber sensing signal.

[0024] Through the above technical solution, by acquiring multiple optical fiber sensing signals; extracting the background signal and intrusion signal of each optical fiber sensing signal; inputting any background signal and any intrusion signal into the preset model for combined reconstruction to generate the corresponding target optical fiber sensing signal; inputting each target optical fiber sensing signal into the optical fiber sensing basic large model to output the signal characteristics corresponding to each target optical fiber sensing signal through the optical fiber sensing basic large model. It can accurately separate useful signals and noise, improve the accuracy of signal recognition, and at the same time enhance the recognition and extraction capabilities of optical fiber sensing signals, reduce the interference and influence of environmental noise, and improve the generalization ability and accuracy of the model.

[0025] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings:

[0027] Figure 1 A schematic diagram of a process of a method for processing optical fiber sensing signals according to an embodiment of the present application is shown;

[0028] Figure 2 A schematic diagram schematically shows a target optical fiber sensing signal synthesis mechanism according to an embodiment of the present application;

[0029] Figure 3 The flowchart of the method for processing optical fiber sensing signals according to a specific embodiment of the present application is schematically shown;

[0030] Figure 4A schematic diagram schematically shows a noise addition-denoising process according to a diffusion model of an embodiment of the present application;

[0031] Figure 5 A schematic diagram of the optical fiber sensing signal separation-synthesis process according to an embodiment of the present application is schematically shown;

[0032] Figure 6 A schematic diagram schematically shows a combined reconstruction of intrusion-background information based on a diffusion model according to an embodiment of the present application;

[0033] Figure 7 The structure block diagram of a device for processing optical fiber sensing signals according to an embodiment of the present application is schematically shown;

[0034] Figure 8 The schematic diagram shows a structure of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0036] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0037] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0038] Figure 1 The following is a schematic diagram of a process flow of a method for processing optical fiber sensing signals according to an embodiment of the present application. Figure 1 As shown, an embodiment of the present application provides a method for processing optical fiber sensing signals, which may include the following steps.

[0039] S102, acquiring multiple optical fiber sensing signals.

[0040] It can be understood that fiber optic sensing signals are a technology that uses optical fiber as a transmission medium to convert the physical quantity of non-optical signals into optical signals for perception and transmission.

[0041] S104, extracting the background signal and the intrusion signal of each optical fiber sensing signal.

[0042] It can be understood that the background signal refers to the signal generated by the environment or system stability, such as soil quality, temperature, pressure, etc. The intrusion signal refers to the signal generated by external factors. Specifically, the background signal and the intrusion signal can be extracted from the real optical fiber sensing signal by feature extraction means.

[0043] In an embodiment of the present application, extracting the background signal and the intrusion signal of each optical fiber sensing signal includes: determining the statistical characteristics of each optical fiber sensing signal; and extracting the background signal and the intrusion signal of each optical fiber sensing signal according to the statistical characteristics.

[0044] Specifically, the statistical features include any one of variance, frequency domain characteristics, local vibration characteristics, and significant regions. Among them, the variance can reflect the energy size and fluctuation of the signal. Background noise is usually a low-frequency, low-variance signal. Intrusion signals (such as abnormal vibrations) often show short-term changes with high variance and high energy. By calculating the local variance of the optical fiber sensing signal, the area with higher variance in the optical fiber sensing signal can be identified, thereby separating possible intrusion signals. Frequency domain characteristics: The time domain signal can be converted to the frequency domain by performing Fourier transform or wavelet transform on the signal, and the spectral characteristics of the vibration signal can be analyzed. Generally speaking, background noise is usually concentrated in the low frequency band, while intrusion signals may appear in a specific frequency range or frequency component. Through filters (such as bandpass filters), the characteristics of intrusion signals can be isolated and extracted. Local vibration characteristics: By analyzing the characteristics of the optical fiber sensing signal in the local area (for example, vibration amplitude and duration), the area significantly different from the background noise can be identified. For intrusion events, vibrations may be sudden and significant, while background noise is usually random and stable. By analyzing the local changes in the vibration waveform (such as peak detection), intrusion events can be distinguished. Significant region detection: Use a significant detection algorithm to identify significant vibration regions in the signal. Significant region detection is a common signal processing technique that uses significant features of the signal (such as abnormal amplitude or frequency components) to identify signal regions that are different from the background. Machine learning models (such as convolutional neural networks) can also be applied to significant region detection to distinguish between background and intrusion information by learning different vibration patterns.

[0045] S106, inputting any background signal and any intrusion signal into a preset model for combined reconstruction to generate a corresponding target optical fiber sensing signal.

[0046] It can be understood that the preset model refers to a model for combining and reconstructing the background signal and the intrusion signal, which can be specifically realized by a diffusion model. The target optical fiber sensing signal is a simulated optical fiber sensing signal generated by combining and reconstructing the background signal and the intrusion signal.

[0047] Specifically, Figure 2As shown, the optical fiber vibration signal is a type of optical fiber sensing signal. For intrusion events, the categories of intrusion events combined with influencing factors, the angle between the optical fiber and the intrusion signal and the influencing factors of the neural network model can be used to obtain the optical fiber vibration signal. Among them, the influencing factors include but are not limited to the laying method of the optical fiber and weather variables. Each intrusion event (such as excavation, drilling, crossing, etc.) will present a different characteristic pattern in the optical fiber sensing signal. Specifically, the features of these patterns can be extracted from the original signal through an algorithm, and then the event classification model can be trained with data from different intrusion events. In this way, when a new signal appears, the model will classify the event according to the features, such as intrusion events such as excavation, drilling or natural interference.

[0048] S108, input each target optical fiber sensing signal into the optical fiber sensing basic large model, so as to output the signal characteristics corresponding to each target optical fiber sensing signal through the optical fiber sensing basic large model.

[0049] It can be understood that the basic model of optical fiber sensing refers to the effective features and information used to identify optical fiber sensing signals. Figure 3 As shown in the figure, a waterfall chart recognition model is constructed based on the CV big model, and a large amount of data on soil quality, environment, and cable burial methods in different regions are used to obtain general feature extraction capabilities. Specifically, buried pipelines cover a variety of soil, temperature, precipitation, burial depth, and environmental noise areas. The deployment of fiber optic sensing equipment can collect a large amount of diverse fiber optic sensing data. By establishing a fiber optic sensing basic big model, a large amount of collected unlabeled data, the B-level parameters of the fiber optic sensing basic big model are pre-trained, and the diversity data will continue to be increased to further enhance the model's capabilities. Based on different buried pipeline scenarios, combined with the fiber optic sensing signals of different buried pipeline scenarios (a small number of examples), the fiber optic sensing basic big model is fine-tuned and verified through the back-end feature extractor, so that the big model has the recognition ability to adapt to different scenarios. After the signal features corresponding to each target fiber optic sensing signal are output through the fiber optic sensing basic big model, the signal features can be input into the oil and gas pipeline application model to classify the intrusion events contained in the target fiber optic sensing signal. For example, excavator, manual, continuous vibration, or normal events.

[0050] Through the above technical solution, by acquiring multiple optical fiber sensing signals; extracting the background signal and intrusion signal of each optical fiber sensing signal; inputting any background signal and any intrusion signal into the preset model for combined reconstruction to generate the corresponding target optical fiber sensing signal; inputting each target optical fiber sensing signal into the optical fiber sensing basic large model to output the signal characteristics corresponding to each target optical fiber sensing signal through the optical fiber sensing basic large model. It can accurately separate useful signals and noise, improve the accuracy of signal recognition, and at the same time enhance the recognition and extraction capabilities of optical fiber sensing signals, reduce the interference and influence of environmental noise, and improve the generalization ability and accuracy of the model.

[0051] In an embodiment of the present application, the preset model is a diffusion model, and the method also includes a step of training the diffusion model: obtaining historical fiber optic sensor signals of historical intrusion events and influencing factors corresponding to the fiber optic sensing; performing noise processing on the historical fiber optic sensor signals according to the influencing factors to obtain corresponding noisy fiber optic sensor signals; performing denoising processing on the noisy fiber optic sensor signals to obtain corresponding denoised fiber optic sensor signals; and training the diffusion model according to the historical fiber optic sensor signals and the denoised fiber optic sensor signals to obtain a trained diffusion model.

[0052] In an embodiment of the present application, a noisy fiber optic sensor signal is denoised to obtain a corresponding denoised fiber optic sensor signal; a main decoupling factor of the noisy fiber optic sensor signal is determined based on a principal component analysis method, the main decoupling factor including background information, intrusion information and noise; and the noisy fiber optic sensor signal is denoised according to the main decoupling factor to obtain a corresponding denoised fiber optic sensor signal.

[0053] In an embodiment of the present application, the influencing factors include at least one of meteorological data, optical fiber laying method, network hanging method, and wind force level.

[0054] Specifically, the preset model is a diffusion model, which can be a diffusion model based on a neural network. The diffusion model training process involves performing denoising and de-noising on real samples under different signal-to-noise ratio conditions. The generation process is to perform denoising multiple times from pure Gaussian noise. The denoising and de-noising process of the diffusion model is as follows: Figure 4 As shown in Figure 2, during the training phase, the model learns to add and remove noise from the data, achieving the transformation from Gaussian distribution to the real data distribution. During the sampling phase, starting from Gaussian noise, the generated samples are obtained through multi-step denoising.

[0055] Specifically, Figure 5 The basic framework of the optical fiber vibration signal synthesis method is schematically shown. First, background information and intrusion information are extracted from the real optical fiber vibration signal by feature extraction. Then, the original signal is reconstructed from the background information and intrusion information by a neural network model NN. When the neural network model NN has the ability to reconstruct the original signal, the generalization ability of the neural network model can ensure that the model can still synthesize a light vibration signal when given unpaired background information and intrusion information. Figure 6 As shown, the process of optical fiber sensing-intrusion information separation-combination generation based on the diffusion model is schematically illustrated.

[0056] In the embodiment of the present application, the functional expression of the objective function of the diffusion model training process is shown in the following formulas (1) and (2):

[0057]

[0058] Among them, t1 refers to the time step, x0 refers to the real sample drawn from the distribution p(x0) of the training data set, and x t It refers to the sample after adding noise to the real sample at time step t1, ∈ θ refers to the preset model, ∈ θ (x t1 ,t1) refers to the predicted output of the preset model at time step t1, ∈ refers to the added noise, Refers to the preset weighting coefficient. Specifically, the added noise can be Gaussian noise, and the time step is a specific stage in the noise addition process. The preset weighting coefficient is used to control the ratio between the original signal and the noise, which changes with the time step. The noise component gradually increases.

[0059] Furthermore, the diffusion model can be fine-tuned using the historical fiber optic sensing signals and the denoised fiber optic sensing signals. θ The input is not only x t and t, as well as the background information of the optical fiber sensor signal x B and intrusion signal x I , the objective function can be optimized and the obtained formula is as follows (4):

[0060]

[0061] Among them, x0 refers to the sample drawn from the training data set distribution p(x0), x t It refers to the sample after adding noise to the real sample at time step t, ∈ θ refers to the preset model, ∈ θ (x t ,t) refers to the predicted output of the preset model at time step t, ∈ refers to the added noise, x B is the background signal, x I It refers to the intrusion signal.

[0062] In the embodiment of the present application, the trained preset model ∈ θ After that, according to the background signal x in different domains B and intrusion signal x I Any combination of can generate a cross-domain combined optical fiber sensing signal sample. Then, any background signal and any intrusion signal are input into the preset model for combined reconstruction to generate the corresponding target optical fiber sensing signal, including the target optical fiber sensing signal is calculated according to the following formula (1):

[0063]

[0064] Among them, t2 refers to the time step, x t2-1It refers to the target optical fiber sensor signal when the time step is t2-1, ∈ θ refers to the preset model, x t2 It refers to the sample after adding noise to the real sample at time step t2, x B is the background signal, x I is the intrusion signal, σ t2 is the noise scale coefficient at time step t2, z is the random variable, a t2 Refers to the preset weighting coefficient, z~N(0,1). The output of the model is used to predict noise and help denoising. The larger the time step, the higher the degree of noise addition and the closer it is to pure noise; the smaller the time step, the closer it is to the original signal. The background signal is the relatively stable part of the fiber optic sensor signal that is not affected by the intrusion. The intrusion signal refers to the signal component caused by an intrusion event (such as construction activities or other external interference). σ t2 Refers to the preset weighting coefficient, which helps the model to gradually remove noise and make the generated model closer to the real signal. Noise scale coefficient controls the proportion of residual noise. The smaller this value is, the closer the generated signal is to the denoised signal. z is a random variable, usually sampled from a standard normal distribution, which is used to introduce a small amount of randomness to make the generated signal diverse.

[0065] Specifically, the diffusion model can realize conditional decoupling and controllable synthesis of optical fiber sensing signals. This is mainly achieved through the following steps:

[0066] Conditional synthesis: Use the diffusion model to analyze the existing sensor signals and synthesize new signal samples according to specific conditions (such as soil quality, intrusion, excavator type, level and other environmental factors). The conditions may include external environmental variables or other intrusion actions.

[0067] Decoupling process: In this step, the diffusion model decouples the complex signal and separates the independent influencing factors. This can be achieved through matrix decomposition or signal processing techniques, for example, using principal component analysis (PCA) to analyze the main decoupling factors in the signal. The main decoupling factors include background information, intrusion information, and noise. The noisy fiber sensing signal is denoised according to the main decoupling factors to obtain the corresponding denoised fiber sensing signal.

[0068] Controllable synthesis: By adjusting the interpolation parameters of the diffusion model decoupling factor, the characteristics of the synthetic signal (such as frequency, amplitude, etc.) can be controlled to generate samples that are difficult to collect in reality or high-value samples.

[0069] Sampling generation: Use controllable sampling methods to generate signal samples that conform to a specific distribution. These samples can be signals from rare events or high-value scenarios, which helps improve the generalization ability and accuracy of the model.

[0070] Data-training closed loop: Synthetic signal samples are used to train machine learning models, which can then be used to more accurately analyze real-world fiber optic sensing data. Through this closed loop, the performance of the model is continuously improved and adapted to more diverse application scenarios.

[0071] The above technical solution can accurately separate useful signals and noise through the diffusion model, improve the accuracy of signal recognition, and enhance the recognition and extraction capabilities of fiber optic sensor signals, reduce the interference and influence of environmental noise, and improve the generalization ability and accuracy of the model. The introduction of the EVIT large model algorithm idea can efficiently process and analyze massive amounts of fiber optic sensor data, greatly reduce computing time and resource consumption, and significantly improve data processing efficiency and model performance.

[0072] In the field of smart agriculture, efficient agricultural machinery monitoring and early warning systems can be used to optimize agricultural production management, reduce false alarms, and improve agricultural production efficiency and safety. In the field of industrial safety, accurate vibration signal recognition and early warning technology can be used to improve the safety management level of industrial facilities, reduce accidents, and improve operating efficiency. In the field of environmental monitoring, it can provide more accurate and real-time monitoring data to help environmental protection and natural disaster prevention. For smart city construction, the technology of the present invention is applied in urban underground pipe networks, subway lines and other areas to improve the intelligent monitoring and management level of urban infrastructure and promote the construction of smart cities. For border and military security, it can be applied to security monitoring of border areas and military facilities, improve the accuracy and real-time nature of intrusion detection, and enhance the security protection capabilities of military facilities. The above technical solution has wide applicability and good application prospects, and can provide more accurate and efficient solutions for various security monitoring and early warning systems, and promote the development and progress of related industries.

[0073] Figure 1 FIG. 1 is a flow chart of a method for processing optical fiber sensing signals in one embodiment. It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0074] Figure 7The structure block diagram of a device for processing optical fiber sensing signals according to an embodiment of the present application is schematically shown. Figure 7 As shown, an embodiment of the present application provides a device for processing optical fiber sensing signals, which may include:

[0075] a memory configured to store instructions; and

[0076] The processor is configured to call instructions from the memory and implement the above-mentioned method for processing optical fiber sensing signals when executing the instructions.

[0077] Specifically, in the embodiment of the present application, the processor may be configured to:

[0078] Acquire multiple optical fiber sensing signals; extract background signals and intrusion signals of each optical fiber sensing signal;

[0079] Input any background signal and any intrusion signal into the preset model for combined reconstruction to generate corresponding target fiber optic sensing signals; input each target fiber optic sensing signal into the fiber optic sensing basic model to output the signal characteristics corresponding to each target fiber optic sensing signal through the fiber optic sensing basic model.

[0080] In an embodiment of the present application, the processor may also be configured to:

[0081] The preset model is a diffusion model, and the method also includes a diffusion model training step: obtaining historical fiber optic sensor signals of historical intrusion events and influencing factors corresponding to fiber optic sensing; performing noise processing on the historical fiber optic sensor signals according to the influencing factors to obtain corresponding noisy fiber optic sensor signals; performing denoising processing on the noisy fiber optic sensor signals to obtain corresponding denoised fiber optic sensor signals; and training the diffusion model according to the historical fiber optic sensor signals and the denoised fiber optic sensor signals to obtain a trained diffusion model.

[0082] In an embodiment of the present application, the processor may also be configured to:

[0083] The noisy fiber optic sensor signal is denoised to obtain the corresponding denoised fiber optic sensor signal; the main decoupling factors of the noisy fiber optic sensor signal are determined based on the principal component analysis method, and the main decoupling factors include background information, intrusion information and noise; the noisy fiber optic sensor signal is denoised according to the main decoupling factors to obtain the corresponding denoised fiber optic sensor signal.

[0084] In an embodiment of the present application, the influencing factors include at least one of meteorological data, optical fiber laying method, network hanging method, and wind force level.

[0085] In the embodiment of the present application, the functional expression of the objective function of the diffusion model training process is shown in the following formulas (1) and (2):

[0086]

[0087] Among them, t1 refers to the time step, x0 refers to the real sample drawn from the distribution p(x0) of the training data set, and x t It refers to the sample after adding noise to the real sample at time step t1, ∈ θ refers to the preset model, ∈ θ (x t1 ,t1) refers to the predicted output of the preset model at time step t1, ∈ refers to the added noise, Refers to the preset weighting coefficient.

[0088] In an embodiment of the present application, the processor may also be configured to:

[0089] Inputting any background signal and any intrusion signal into the preset model for combined reconstruction to generate the corresponding target optical fiber sensing signal includes: the target optical fiber sensing signal is calculated according to the following formula (3):

[0090]

[0091] Among them, t2 refers to the time step, x t2-1 It refers to the target optical fiber sensor signal when the time step is t2-1, ∈ θ refers to the preset model, x t2 It refers to the sample after adding noise to the real sample at time step t2, x B is the background signal, x I is the intrusion signal, σ t2 is the noise scale coefficient at time step t2, z is the random variable, a t2 Refers to the preset weighting coefficient.

[0092] In an embodiment of the present application, the processor may also be configured to:

[0093] Extracting the background signal and the intrusion signal of each optical fiber sensing signal includes: determining the statistical characteristics of each optical fiber sensing signal; and extracting the background signal and the intrusion signal of each optical fiber sensing signal according to the statistical characteristics.

[0094] In the embodiment of the present application, the statistical feature includes any one of variance, frequency domain characteristics, local vibration characteristics, and significant area.

[0095] An embodiment of the present application further provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the above-mentioned method for processing optical fiber sensing signals.

[0096] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor A01, a network interface A02, a memory (not shown in the figure) and a database (not shown in the figure) connected through a system bus. Among them, the processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01, a computer program B02 and a database (not shown in the figure). The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A04. The database of the computer device is used to store processing data of optical fiber sensor signals. The network interface A02 of the computer device is used to communicate with an external terminal through a network connection. When the computer program B02 is executed by the processor A01, a method for processing optical fiber sensor signals is implemented.

[0097] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0098] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0099] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0100] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0101] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0102] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0103] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0104] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. 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 (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0105] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0106] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for processing optical fiber sensing signals, characterized in that: include: Acquire multiple optical fiber sensing signals; Extracting background signal and intrusion signal of each optical fiber sensing signal; Input any background signal and any intrusion signal into the preset model for combined reconstruction to generate the corresponding target optical fiber sensing signal; Each target optical fiber sensing signal is input into the optical fiber sensing basic large model, so as to output the signal characteristics corresponding to each target optical fiber sensing signal through the optical fiber sensing basic large model.

2. The method for processing optical fiber sensing signals according to claim 1, characterized in that: The preset model is a diffusion model, and the method further comprises a step of training the diffusion model: Obtain historical fiber optic sensing signals of historical intrusion events and influencing factors corresponding to fiber optic sensing; Performing noise processing on the historical optical fiber sensing signal according to the influencing factors to obtain a corresponding noisy optical fiber sensing signal; De-noising the noisy optical fiber sensing signal to obtain a corresponding de-noised optical fiber sensing signal; The diffusion model is trained according to the historical optical fiber sensing signal and the denoised optical fiber sensing signal to obtain a trained diffusion model.

3. The method for processing optical fiber sensing signals according to claim 2, characterized in that: The denoising process is performed on the noisy optical fiber sensing signal to obtain a corresponding denoised optical fiber sensing signal; Determine the main decoupling factors of the noisy optical fiber sensing signal based on a principal component analysis method, wherein the main decoupling factors include background information, intrusion information and noise; The noise optical fiber sensing signal is denoised according to the main decoupling factor to obtain a corresponding denoised optical fiber sensing signal.

4. The method for processing optical fiber sensing signals according to claim 2, characterized in that: The influencing factors include at least one of meteorological data, optical fiber laying method, network hanging method, and wind force level.

5. The method for processing optical fiber sensing signals according to claim 2, characterized in that: The functional expression of the objective function of the diffusion model training process is shown in the following formulas (1) and (2): Among them, t1 refers to the time step, x0 refers to the real sample drawn from the distribution p(x0) of the training data set, and x t It refers to the sample after adding noise to the real sample at time step t1, ∈ θ refers to the preset model, ∈ θ (x t1 ,t1) refers to the predicted output of the preset model at time step t1, ∈ refers to the added noise, Refers to the preset weighting coefficient.

6. The method for processing optical fiber sensing signals according to claim 1, characterized in that: The step of inputting any background signal and any intrusion signal into a preset model for combined reconstruction to generate a corresponding target optical fiber sensing signal includes calculating the target optical fiber sensing signal according to the following formula (1): Among them, t2 refers to the time step, x t2-1 It refers to the target optical fiber sensor signal when the time step is t2-1, ∈ θ refers to the preset model, x t2 It refers to the sample after adding noise to the real sample at time step t2, x B is the background signal, x I is the intrusion signal, σ t2 is the noise scale coefficient at time step t2, z is the random variable, a t2 Refers to the preset weighting coefficient.

7. The method for processing optical fiber sensing signals according to claim 1, characterized in that: The extracting of the background signal and the intrusion signal of each optical fiber sensing signal comprises: determining statistical characteristics of each optical fiber sensing signal; The background signal and the intrusion signal of each optical fiber sensing signal are extracted according to the statistical characteristics.

8. The method for processing optical fiber sensing signals according to claim 7, characterized in that: The statistical features include any one of variance, frequency domain characteristics, local vibration characteristics, and significant regions.

9. A device for processing optical fiber sensing signals, characterized in that: include: a memory configured to store instructions; A processor is configured to call the instruction from the memory and implement the method for processing the optical fiber sensing signal according to any one of claims 1 to 8 when executing the instruction.

10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for causing a machine to execute the processing of the optical fiber sensing signal according to any one of claims 1 to 8.

Images