Environment detection system and method based on fiber grating sensor

Abstract

The invention discloses an environment detection system and method based on a fiber bragg grating sensor in the technical field of fiber bragg grating application, and the method comprises the steps: detecting an environment through the fiber bragg grating sensor, and obtaining a grating reflection signal; stratified analysis is carried out on the grating reflection signal pair according to the physical variable type, and different physical variable information is extracted; calculating the coupling strength between different pieces of physical variable information; performing characteristic adjustment and decoupling control on the fiber grating sensor based on the coupling strength; and performing nonlinear error compensation on the physical variable information after decoupling control by using an error compensation model. Through combination of the fiber grating sensor and the error compensation model, deep fusion of a physical layer and an algorithm layer is realized, so that the system can better deal with various interference factors in a complex environment, and the fiber grating sensor can perform accurate measurement in the complex environment.

Description

Technical Field

[0001] The invention relates to an environment detection system and method based on a fiber grating sensor, belonging to the technical field of fiber grating application. Background Art

[0002] Fiber Bragg grating sensor is a type of fiber optic sensor. The sensing process based on fiber grating is to obtain sensing information by modulating the Bragg wavelength of the optical fiber by external physical parameters. It is a wavelength modulated fiber optic sensor.

[0003] Fiber Bragg grating sensors are widely used in engineering monitoring, structural health monitoring and other fields due to their high sensitivity and anti-electromagnetic interference characteristics. However, existing technologies still face problems in practical applications, such as incomplete compensation for environmental interference, difficulty in decoupling temperature and strain, difficulty in accurately compensating nonlinear errors, and complex signal processing. These problems are interrelated and hinder the accurate measurement of fiber Bragg grating sensors in complex environments. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide an environmental detection system and method based on a fiber grating sensor. By combining the fiber grating sensor with an error compensation model, the problem of multiple interference factors in a complex environment is solved, and the fiber grating sensor can be accurately measured in a complex environment.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] In a first aspect, the present invention provides an environment detection method based on a fiber grating sensor, comprising:

[0007] Use fiber grating sensor to detect the environment and obtain grating reflection signal;

[0008] Perform hierarchical analysis on the grating reflection signal pairs according to the type of physical variables to extract information of different physical variables;

[0009] Calculate the coupling strength between different physical variable information;

[0010] Characteristic adjustment and decoupling control of fiber Bragg grating sensors based on coupling strength;

[0011] The error compensation model is used to perform nonlinear error compensation on the physical variable information after decoupling control.

[0012] Furthermore, the fiber grating sensor integrates multimode fiber gratings with specific sensitivity to different physical variables in the same sensing node, and the multimode fiber gratings are arranged at different angles and intervals to perform differentiated detection on multiple physical variables.

[0013] Furthermore, the decoupling control formula includes:

[0014] Δλ=k T ΔT+k S ΔS

[0015]

[0016] Where: Δλ is the change in reflected wavelength, k T and k S is the coefficient of the effect of temperature and strain on wavelength, ΔT and ΔS are the changes of temperature and strain, u(t) is the control variable, K p , K i , K d are proportional, integral and differential gains respectively, and e(t) is the error signal.

[0017] Furthermore, the formula of the error compensation model includes:

[0018] y=σ(W n ·σ(W n-1 ·(...σ(W2·σ(W1x+b1)+b2)...)+b n ))

[0019]

[0020] Where: y is the output of the error compensation model, σ is the activation function, W i and b i are the weight and bias of the i-th layer, x is the strain signal, L is the loss function, N is the number of samples, is the predicted value of the model, y i is the actual value.

[0021] 6. Furthermore, the fiber grating sensor uses the intelligent processing module of the edge node to preprocess the collected grating reflection signal, and the preprocessing formula includes:

[0022] p=IDWT(W·x)

[0023] Where W represents the wavelet coefficient, IDWT is the inverse wavelet transform, x is the input signal, and p is the optimized signal;

[0024] The preprocessed data is transmitted to the cloud for deep correlation analysis, and a global environmental response model is constructed for parameter correction and then fed back to the edge node. In addition, the collaborative optimization algorithm between the edge and the cloud is used to generate a parameter optimization strategy based on the overall distribution of historical data, and the parameter optimization strategy is adjusted through real-time feedback from the edge node.

[0025] The formula for collaborative optimization between edge and cloud includes:

[0026] θ new =θ old -α▽L(θ)

[0027] Among them, θ old is the model parameter before optimization, θ new are the optimized model parameters, α is the learning rate, and ▽L(θ) is the gradient of the loss function.

[0028] Furthermore, the fiber Bragg grating sensor utilizes an adaptive threshold detection mechanism to perform dynamic multi-stage calibration and self-calibration on the global environmental response model. The dynamic multi-stage calibration includes initial calibration, environmental switching calibration and dynamic self-calibration. The self-calibration includes setting multiple reference points of known physical quantities to continuously calibrate and adjust during operation. The calibration adjustment formula includes:

[0029] θ adjust =κ(θ ref -θ meas )

[0030] Among them, θ ref is the ideal measurement value of the reference point, θ meas is the actual measured value, θ adjust is the calibration parameter adjustment, and κ is the calibration coefficient.

[0031] Furthermore, the sensor nodes of the fiber grating sensor are interconnected through a network to form a collaborative sensing network, which uses the collaborative sensing network to share information and work in collaboration with other nodes, and adjusts the measurement parameters and correction strategies of the sensor node in real time through a distributed optimization strategy based on the measurement data of other sensor nodes.

[0032] In a second aspect, the present invention provides an environment detection system based on a fiber grating sensor, comprising:

[0033] Signal detection module: used to detect the environment using a fiber grating sensor to obtain a grating reflection signal;

[0034] Variable extraction module: used to perform hierarchical analysis on grating reflection signal pairs according to the type of physical variables and extract information of different physical variables;

[0035] Variable calculation module: used to calculate the coupling strength between different physical variable information;

[0036] Decoupling control module: used to adjust the characteristics and perform decoupling control on the fiber Bragg grating sensor based on the coupling strength;

[0037] Error compensation module: Use the error compensation model to perform nonlinear error compensation on the physical variable information after decoupling control.

[0038] In a third aspect, the present invention provides an environment detection device based on a fiber grating sensor, including a processor and a storage medium;

[0039] The storage medium is used to store instructions;

[0040] The processor is used to operate according to the instructions to execute the steps of any of the methods described above.

[0041] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any of the above methods when executed by a processor.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. Collaborative innovation of the physical layer and the algorithm layer: The present invention realizes the deep integration of the physical layer and the algorithm layer by combining the fiber Bragg grating sensor with the error compensation model, so that the system can better cope with various interference factors in complex environments;

[0044] 2. Architecture design of intelligent edge computing and cloud collaboration: The present invention realizes the optimal configuration of computing resources by combining edge computing and cloud computing, ensuring the real-time performance and high precision of the system in complex environments;

[0045] 3. Introduction of dynamic multi-stage calibration and self-calibration mechanism: Through dynamic calibration and self-calibration strategies, the system can automatically maintain high accuracy under various environmental conditions, reducing errors caused by environmental changes;

[0046] 4. Construction of intelligent collaborative perception network: The application of multi-node collaborative work and distributed optimization algorithm in the present invention realizes mutual compensation and global optimization among multi-grating systems, and improves the stability and measurement accuracy of the overall system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0048] Figure 1 A schematic flow chart of an environmental detection method based on a fiber grating sensor provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0049] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present invention can be combined with each other without conflict.

[0050] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present invention belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present invention.

[0051] Embodiment 1:

[0052] This scheme adopts an environmental detection method based on fiber grating sensors, aiming to effectively decouple temperature and strain from a physical level and achieve accurate compensation for multiple environmental factors. The intelligent multi-mode collaborative sensing structure used in this environmental detection method based on fiber grating sensors adopts a multi-mode grating integrated design: by integrating multi-mode fiber gratings (such as long-period gratings combined with Bragg gratings) in the same sensing node, these gratings have specific sensitivities to different physical quantities, and can achieve differentiated detection of multiple variables such as temperature and strain. Long-period gratings are highly sensitive to ambient temperature, while Bragg gratings are highly sensitive to strain. Through clever physical arrangement and integrated design, separate perception of temperature and strain can be achieved.

[0053] In addition, this solution optimizes the arrangement of the integrated gratings, arranging the long-period grating and the Bragg grating at different angles and intervals to reduce the mutual influence between them. Specifically, the long-period grating can be set at the periphery of the sensing node to capture the ambient temperature, while the Bragg grating is set at the center to focus on sensing the strain. In this way, through reasonable spatial arrangement, the coupling effect between temperature and strain signals can be effectively reduced.

[0054] In terms of physical structure, this solution realizes the physical decoupling of temperature and strain through the integrated design of multi-mode gratings, and further realizes hierarchical analysis in the signal processing part. By independently collecting data and extracting features of the reflection wavelength changes of each grating, the system can effectively distinguish temperature and strain information, avoid mutual interference, and improve the accuracy and reliability of data.

[0055] In this solution, an adaptive coupling eliminator is used to actively control the interaction between gratings and adjust the reflection characteristics of different gratings in real time, thereby reducing the mutual interference between different physical quantities. This active coupling elimination mechanism is different from the traditional static calibration method. It is controlled through a dynamic closed-loop feedback system and can significantly improve the decoupling accuracy of temperature and strain.

[0056] The closed-loop feedback system consists of an intelligent control module and an actuator on the sensor node. By real-time monitoring of the reflected signal of the grating, the feedback system can calculate the coupling strength between temperature and strain, and drive the actuator to adjust the characteristics of the grating, such as changing the tensile state of the optical fiber or the local temperature, to achieve real-time coupling elimination. The decoupling control equation is:

[0057] Assuming that the change in reflection wavelength is Δλ, and the changes in temperature and strain are ΔT and ΔS respectively, the response model of the sensor is:

[0058] Δλ=k T ΔT+k S ΔS

[0059] Among them, k T and k S It is the coefficient of the influence of temperature and strain on wavelength.

[0060] Through feedback control, the control variable u is adjusted to minimize the coupling error e(t):

[0061]

[0062] Among them, K p , K i , K d are proportional, integral and differential gains respectively, and e(t) is the error signal.

[0063] This solution aims to solve the nonlinear error problem that cannot be effectively handled in the existing system. It proposes to introduce deep learning technology and establish an intelligent error compensation model to effectively compensate for the complex nonlinear behavior of the system. Specifically, it includes:

[0064] Data-driven error learning model: Use deep neural networks (DNN) to build an error compensation model for fiber Bragg grating sensors. By collecting a large amount of experimental data under different conditions, a deep learning model that can describe the nonlinear response of temperature, strain, and other environmental variables is trained. By capturing the complex characteristics of the system, the model can accurately compensate for the errors of the measured data in real-time applications.

[0065] Nonlinear error compensation formula: Assuming that the input data is the temperature and strain signal x, and the output is the compensated error y, the output of the deep neural network model can be expressed as:

[0066] y=σ(W n ·σ(W n-1 ·(...σ(W2·σ(W1x+b1)+b2)...)+b n ))

[0067] Among them, σ is the activation function, W iand b i are the weight and bias of the i-th layer respectively.

[0068] By minimizing the loss function (such as mean square error MSE), the parameters of the model can be optimized to achieve the best error compensation effect of the system:

[0069]

[0070] in, is the predicted value of the model, y i is the actual value, and N is the number of samples.

[0071] Online learning and continuous optimization: Deep learning models not only rely on initial training data, but can also perform online learning and continuous optimization during the long-term operation of the sensor. By comparing the actual measurement data of the sensor with the benchmark data, the model can continuously adjust its internal parameters to adapt to the state changes of the sensor and the dynamic changes of the environment, thereby maintaining long-term stable high-precision measurement.

[0072] In order to meet the real-time requirements of complex signal processing, this solution designs an intelligent edge computing and cloud collaborative architecture, which achieves the optimal balance of the overall system performance by distributing some data processing tasks to edge computing nodes and utilizing the high-performance computing capabilities of the cloud.

[0073] This solution deploys a lightweight intelligent processing module at the edge node of the sensor. The module is supported by a dedicated integrated circuit (such as FPGA or ASIC) and can pre-process the collected grating reflection signal, including denoising, feature extraction, and preliminary error compensation. The intelligent processing module can effectively reduce the bandwidth pressure caused by the original data transmission and improve the response speed of the system; the edge node filters and extracts features of the input signal x, and uses wavelet transform to denoise the signal. The edge node optimization formula is:

[0074] p=IDWT(W·x)

[0075] Where W represents the wavelet coefficient, IDWT is the inverse wavelet transform, x is the input signal, and p is the optimized signal. This process can effectively remove noise and extract important features from the signal, thereby reducing the complexity of subsequent calculations.

[0076] The data processed by the edge node is transmitted to the cloud, which performs further in-depth analysis and optimization. The high-performance computing platform in the cloud can perform in-depth correlation analysis on the data of multiple nodes, build a global environmental response model, and feed back the corrected parameters to the edge node to ensure real-time and accuracy.

[0077] By designing an optimization algorithm for edge-cloud collaboration, the cloud can generate parameter optimization strategies based on the overall distribution of historical data, and adjust the parameter optimization strategies through real-time feedback from edge nodes. The collaborative optimization formula for edge-cloud collaboration is as follows:

[0078] θ new =θ old -α▽L(θ)

[0079] Among them, θ old is the model parameter before optimization, θ new are the optimized model parameters, α is the learning rate, and ▽L(θ) is the gradient of the loss function. The edge node further improves the measurement accuracy and real-time response capability by applying the optimized parameters sent back by the cloud.

[0080] In order to solve the problem of calibration model failure in existing systems, this solution introduces a dynamic multi-stage calibration and self-calibration mechanism to ensure the reliability and stability of the system under long-term and multi-environment conditions. The system calibration process is divided into multiple stages, including initial calibration, environment switching calibration and dynamic self-calibration, where:

[0081] Initial calibration: Establish the basic response model of the sensor through high-precision benchmark measurements under laboratory conditions. This stage of calibration ensures that the system has the best measurement performance under ideal conditions and is the benchmark for subsequent stages.

[0082] Environmental switch calibration: When the system detects a significant change in the environment (such as a sudden change in temperature or a significant change in the stress environment), the calibration procedure is automatically triggered. This stage aims to ensure the accuracy of the measurement results by adjusting the model parameters to adapt to the new environmental conditions.

[0083] Dynamic self-calibration: During the actual operation of the system, by using real-time measurement data to make small self-corrections, the system can cope with gradually accumulated environmental changes and hardware aging factors, maintaining the continued applicability of the model.

[0084] In addition, the system is equipped with an adaptive threshold detection mechanism to monitor the degree of environmental change in real time. Once a change beyond the set threshold is detected, the environment switch calibration is immediately triggered to ensure the effectiveness and accuracy of the model. This mechanism ensures that the system can respond quickly and recalibrate in different environments to adapt to changing environments.

[0085] The system is also equipped with a self-calibration mechanism based on reference points. During the actual operation of the system, by setting multiple reference points of known physical quantities (such as temperature reference points or known strain points), the system can continuously verify and adjust its own measurement results during operation. This self-calibration mechanism is based on the principle of feedback control. By regularly comparing the measured values ​​of the reference points with the standard values, the system parameters are corrected, thereby effectively reducing long-term drift and measurement deviation. The calibration adjustment equation includes:

[0086] Assume that the ideal measurement of the reference point is θ ref , the actual measured value is θ meas , then the calibration parameter adjustment θ adjust for:

[0087] θ adjust =κ(θ ref -θ meas )

[0088] Among them, κ is the calibration coefficient, and the system gradually approaches the ideal state through feedback adjustment.

[0089] In order to solve the problem of mutual influence in multi-grating system, this scheme proposes an intelligent collaborative perception network, which optimizes the overall system performance through the collaborative work between multiple sensors. Each sensor node is connected to each other through a wireless network or a wired network to form a collaborative perception network. In this network, each node is not just an independent measurement unit, but works with other nodes by sharing information. The system uses a distributed optimization algorithm to adjust the measurement parameters and correction strategies of each node in real time according to the measurement data of each node, thereby reducing the overall error caused by optical signal crosstalk and environmental factors.

[0090] Each sensor node adjusts the measurement parameters through a distributed optimization strategy and solves based on the Lagrange multiplier method. Each node can optimize its own parameters independently and share the results through the network to achieve global optimization. Specifically, it includes:

[0091] Multi-node collaborative working mechanism: By building a multi-node collaborative working mechanism, the system can perform real-time data smoothing and compensation under different working conditions to ensure the accuracy of the overall measurement results. While completing its own measurement, each node is also responsible for receiving and analyzing the data of surrounding nodes to form a dynamic measurement network to cope with rapidly changing environments.

[0092] Distributed data fusion and feedback control: In the collaborative sensing network, the system uses a distributed data fusion algorithm to comprehensively analyze data from different nodes to establish a global environmental response model. Through the feedback control mechanism, the correction parameters of each node can be dynamically adjusted to adapt to changes in the overall environment and ensure the accuracy and consistency of measurements across the entire network.

[0093] Data fusion and intelligent decision-making: The system integrates the measurement data of each node through a distributed data fusion algorithm to form an intelligent assessment of the overall environment. For example, the system can generate a real-time environmental status map by fusing the temperature and strain data of multiple nodes, and make intelligent decisions based on this to optimize the measurement strategy of each node.

[0094] Network redundancy and fault-tolerance mechanism: In the collaborative sensing network, network redundancy and fault-tolerance mechanism are designed to improve the reliability of the system. Even if some nodes fail to work due to failure, other nodes can still complete the global sensing task through information sharing and redundant calculation, ensuring the stable operation of the system and the continuity of the measurement results.

[0095] Embodiment 2:

[0096] An environment detection system based on a fiber Bragg grating sensor can implement an environment detection method based on a fiber Bragg grating sensor as described in Embodiment 1, comprising:

[0097] Signal detection module: used to detect the environment using a fiber grating sensor to obtain a grating reflection signal;

[0098] Variable extraction module: used to perform hierarchical analysis on grating reflection signal pairs according to the type of physical variables and extract information of different physical variables;

[0099] Variable calculation module: used to calculate the coupling strength between different physical variable information;

[0100] Decoupling control module: used to adjust the characteristics and perform decoupling control on the fiber Bragg grating sensor based on the coupling strength;

[0101] Error compensation module: Use the error compensation model to perform nonlinear error compensation on the physical variable information after decoupling control.

[0102] Embodiment three:

[0103] The embodiment of the present invention further provides an environment detection device based on a fiber Bragg grating sensor, which can implement an environment detection method based on a fiber Bragg grating sensor described in the first embodiment, including a processor and a storage medium;

[0104] The storage medium is used to store instructions;

[0105] The processor is used to operate according to the instructions to perform the steps of the following method:

[0106] Use fiber grating sensor to detect the environment and obtain grating reflection signal;

[0107] Perform hierarchical analysis on the grating reflection signal pairs according to the type of physical variables to extract information of different physical variables;

[0108] Calculate the coupling strength between different physical variable information;

[0109] Characteristic adjustment and decoupling control of fiber Bragg grating sensors based on coupling strength;

[0110] The error compensation model is used to perform nonlinear error compensation on the physical variable information after decoupling control.

[0111] Embodiment 4:

[0112] The embodiment of the present invention further provides a computer-readable storage medium, which can implement the environmental detection method based on the fiber grating sensor described in the first embodiment, and stores a computer program thereon, which implements the steps of the following method when the program is executed by a processor:

[0113] Use fiber grating sensor to detect the environment and obtain grating reflection signal;

[0114] Perform hierarchical analysis on the grating reflection signal pairs according to the type of physical variables to extract information of different physical variables;

[0115] Calculate the coupling strength between different physical variable information;

[0116] Characteristic adjustment and decoupling control of fiber Bragg grating sensors based on coupling strength;

[0117] The error compensation model is used to perform nonlinear error compensation on the physical variable information after decoupling control.

[0118] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

[0119] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take 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.) containing computer-usable program code.

[0120] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of 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 produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. 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.

[0121] 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.

[0122] 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.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for environmental detection based on fiber grating sensor, characterized in that: include: Use fiber grating sensor to detect the environment and obtain grating reflection signal; Perform hierarchical analysis on the grating reflection signal pairs according to the type of physical variables to extract information of different physical variables; Calculate the coupling strength between different physical variable information; Characteristic adjustment and decoupling control of fiber Bragg grating sensors based on coupling strength; The error compensation model is used to perform nonlinear error compensation on the physical variable information after decoupling control.

2. The environmental detection method based on fiber grating sensor according to claim 1 is characterized in that: The fiber grating sensor integrates multi-mode fiber gratings with specific sensitivity to different physical variables in the same sensing node, and the multi-mode fiber gratings are arranged at different angles and intervals to perform differentiated detection on multiple physical variables.

3. The environmental detection method based on fiber grating sensor according to claim 1 is characterized in that: The decoupling control formula includes: Δλ=k T ΔT+k S ΔS Where: Δλ is the change in reflected wavelength, k T and k S is the coefficient of the effect of temperature and strain on wavelength, ΔT and ΔS are the changes of temperature and strain, u(t) is the control variable, K p , K i , K d are proportional, integral and differential gains respectively, and e(t) is the error signal.

4. The environmental detection method based on fiber grating sensor according to claim 1 is characterized in that: The formula of the error compensation model includes: y=σ(W n ·σ(W n-1 ·(...σ(W2·σ(W1x+b1)+b2)...)+b n )) Where: y is the output of the error compensation model, σ is the activation function, W i and b i are the weight and bias of the i-th layer, x is the strain signal, L is the loss function, N is the number of samples, is the predicted value of the model, y i is the actual value.

5. The environmental detection method based on fiber grating sensor according to claim 1 is characterized in that: The fiber grating sensor uses the intelligent processing module of the edge node to preprocess the collected grating reflection signal. The preprocessing formula includes: p=IDWT(W·x) Where W represents the wavelet coefficient, IDWT is the inverse wavelet transform, x is the input signal, and p is the optimized signal; The preprocessed data is transmitted to the cloud for deep correlation analysis, and a global environmental response model is constructed for parameter correction and then fed back to the edge node. In addition, the collaborative optimization algorithm between the edge and the cloud is used to generate a parameter optimization strategy based on the overall distribution of historical data, and the parameter optimization strategy is adjusted through real-time feedback from the edge node. The formula for collaborative optimization between edge and cloud includes: Among them, θ old is the model parameter before optimization, θ new is the optimized model parameter, α is the learning rate, is the gradient of the loss function.

6. The environmental detection method based on fiber grating sensor according to claim 5 is characterized in that: The fiber Bragg grating sensor uses an adaptive threshold detection mechanism to perform dynamic multi-stage calibration and self-calibration on the global environmental response model. The dynamic multi-stage calibration includes initial calibration, environmental switching calibration and dynamic self-calibration. The self-calibration includes setting multiple reference points of known physical quantities to continuously calibrate and adjust during operation. The calibration adjustment formula is include: i adjust =κ(θ ref -θ meas ) Among them, θ ref is the ideal measurement value of the reference point, θ meas is the actual measured value, θ adjust is the calibration parameter adjustment, and κ is the calibration coefficient.

7. The environmental detection method based on fiber grating sensor according to claim 1 is characterized in that: The sensor nodes of the fiber grating sensor are interconnected through a network to form a collaborative sensing network, which uses the collaborative sensing network to share information and work in collaboration with other nodes, and adjusts the measurement parameters and correction strategies of the sensor node in real time through a distributed optimization strategy based on the measurement data of other sensor nodes.

8. An environmental detection system based on fiber grating sensor, characterized in that: include: Signal detection module: used to detect the environment using a fiber grating sensor to obtain a grating reflection signal; Variable extraction module: used to perform hierarchical analysis on grating reflection signal pairs according to the type of physical variables and extract information of different physical variables; Variable calculation module: used to calculate the coupling strength between different physical variable information; Decoupling control module: used to adjust the characteristics and perform decoupling control on the fiber Bragg grating sensor based on the coupling strength; Error compensation module: Use the error compensation model to perform nonlinear error compensation on the physical variable information after decoupling control.

9. An environmental detection device based on a fiber grating sensor, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images