A Method and System for Multi-parameter Detection of Seawater Based on Fiber Optic Whispering Galerie Mode

Abstract

This invention discloses a method and system for multi-parameter detection of seawater based on fiber optic whispering wall mode. This invention relates to the technical field of multi-parameter detection. In the process of multi-parameter detection of seawater, a theoretical analysis model of the fiber optic microbubble cavity whispering wall mode is acquired, and numerical simulation is performed using multiphysics finite element analysis. The influence of the cavity structure parameters of the microbubble cavity on the quality factor is determined based on the iteration of multiple simulation results. A multiphysics coupling model of the microbubble cavity is established to simulate the temperature gradient field and stress distribution of seawater during transmission within the microbubble cavity, improving the accuracy of the results for various parameter changes in seawater. Based on the results of each parameter change, the spectral changes of seawater based on the fiber optic whispering wall mode are determined; the corresponding spectral shifts are marked; and multiple basic parameters are determined by tracing along these spectral shifts. Based on these basic parameters, the comprehensive detection result of seawater under multiple parameters is determined, improving the accuracy of the comprehensive detection result.

Description

Technical Field

[0001] This invention relates to the technical field of multi-parameter detection, and in particular to a method and system for multi-parameter detection of seawater based on fiber optic whispering gallery mode. Background Technology

[0002] Multi-parameter monitoring of seawater is fundamental to studying ocean dynamics, hydrological information, and marine resource development. Among existing seawater monitoring technologies, fiber optic whispering-gallery (WGM) microcavity sensors are widely used for detecting parameters such as seawater temperature, salinity, and depth due to their advantages of ultra-high sensitivity, low loss, and resistance to electromagnetic interference.

[0003] Existing technologies typically analyze the relationship between the cavity structural parameters (such as diameter and wall thickness) and the quality factor (Q value) of microbubble chambers in isolation, lacking in-depth correlation analysis based on iterative optimization. They often ignore the dynamic influence mechanism of microbubble chamber structural parameters on the quality factor, which makes it impossible to guarantee the accuracy of the results of changes in various parameters when simulating the transmission process of seawater inside microbubble chambers, resulting in low accuracy of the comprehensive detection results. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a method and system for detecting multiple parameters of seawater based on fiber optic whispering wall mode.

[0005] This invention provides a method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode, comprising:

[0006] In the process of multi-parameter detection of seawater, a theoretical analysis model of the whispering gallery mode of fiber optic microbubble cavity is collected, and numerical simulation is carried out in combination with multiphysics finite element analysis. Multiple simulation results are output, and the influence relationship between the cavity structure parameters of microbubble cavity and the quality factor is determined based on the iteration of multiple simulation results. The influence of the cavity structure parameters of the microbubble cavity on the quality factor was marked, and a multi-physics coupling model of the microbubble cavity was established. In the multi-physics coupling model of the microbubble cavity, the temperature gradient field and stress distribution during seawater transport in the microbubble cavity were simulated, and the changes of various parameters of the seawater were marked. The spectral changes of seawater based on the fiber optic whispering gallery mode are determined based on the results of the changes in various parameters; the spectral changes are analyzed and the corresponding spectral shifts are marked. By tracing the spectral shift, several fundamental parameters are determined, and based on these parameters, a comprehensive detection result of seawater under these parameters is determined.

[0007] This invention provides a multi-parameter seawater detection system based on fiber optic whispering wall mode. This system is applied to the aforementioned multi-parameter seawater detection method based on fiber optic whispering wall mode. The multi-parameter seawater detection system based on fiber optic whispering wall mode includes: The simulation module is used to acquire the theoretical analysis model of the whispering wall mode of fiber optic microbubble cavity during the multi-parameter detection process of seawater, and to perform numerical simulation by combining multiphysics finite element analysis, outputting multiple simulation results, and determining the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor based on the iteration of multiple simulation results. The parameter variation module is used to mark the influence of the cavity structure parameters of the microbubble cavity on the quality factor, and to establish a multi-physics coupling model of the microbubble cavity. In the multi-physics coupling model of the microbubble cavity, the temperature gradient field and stress distribution of seawater during transport in the microbubble cavity are simulated, and the results of various parameter variations of seawater are marked. The spectral variation module is used to determine the spectral variation of seawater based on the fiber optic whispering gallery mode according to the variation results of various parameters; analyze the spectral variation and mark the corresponding spectral shift. The integrated detection module is used to determine multiple basic parameters by tracing along the spectral shift, and to determine the integrated detection results of seawater under multiple parameters based on these basic parameters.

[0008] Compared with the prior art, the beneficial effects of the present invention are: (1) In the process of multi-parameter detection of seawater, the theoretical analysis model of the whispering wall mode of fiber optic microbubble cavity is collected, and numerical simulation is carried out in combination with multi-physics finite element analysis. Multiple simulation results are output. The influence relationship between the cavity structure parameters of microbubble cavity and the quality factor is determined based on the iteration of multiple simulation results. The influence relationship between the cavity structure parameters of microbubble cavity and the quality factor is marked. A multi-physics coupling model of microbubble cavity is established. In the multi-physics coupling model of microbubble cavity, the temperature gradient field and stress distribution of seawater are simulated when it is transmitted in microbubble cavity. The results of the change of various parameters of seawater are marked. The influence relationship between the cavity structure parameters of microbubble cavity and the quality factor is introduced. The multi-physics coupling model of microbubble cavity is further controlled, which improves the simulation effect of temperature gradient field and stress distribution, and further improves the accuracy of the results of the change of various parameters.

[0009] (2) Determine the spectral changes of seawater based on the fiber optic whispering wall mode according to the results of the changes of each parameter; analyze the spectral changes and mark the corresponding spectral shifts; determine multiple basic parameters by tracing the spectral shifts; determine the comprehensive detection results of seawater under multiple parameters based on the multiple basic parameters; further control the spectral shifts and improve the accuracy of multiple basic parameters and the accuracy of the comprehensive detection results. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the multi-parameter detection method for seawater based on fiber optic whispering gallery mode in an embodiment of the present invention. Figure 2 This is a flowchart illustrating step S11 in the seawater multi-parameter detection method based on fiber optic whispering gallery mode in an embodiment of the present invention. Figure 3 This is a flowchart illustrating step S12 in the seawater multi-parameter detection method based on fiber optic whispering wall mode in an embodiment of the present invention. Figure 4 This is a flowchart illustrating step S13 in the seawater multi-parameter detection method based on fiber optic whispering wall mode in an embodiment of the present invention. Figure 5 This is a flowchart illustrating step S14 in the seawater multi-parameter detection method based on fiber optic whispering wall mode in an embodiment of the present invention. Figure 6 This is a schematic diagram of the structural composition of a seawater multi-parameter detection system based on fiber optic whispering gallery mode in an embodiment of the present invention. Detailed Implementation

[0011] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0012] Please see Figures 1 to 6 A multi-parameter detection method for seawater based on fiber optic whispering gallery mode is proposed and applied to multi-parameter detection scenarios. The multi-parameter detection method for seawater based on fiber optic whispering gallery mode includes: Step S11: During the multi-parameter detection of seawater, a theoretical analysis model of the whispering wall mode of the fiber optic microbubble cavity is collected, and numerical simulation is performed in combination with multi-physics finite element analysis. Multiple simulation results are output, and the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor is determined based on the iteration of multiple simulation results. Step S12: Mark the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor, and establish a multi-physics coupling model of the microbubble cavity. In the multi-physics coupling model of the microbubble cavity, simulate the temperature gradient field and stress distribution when seawater is transported in the microbubble cavity, and mark the changes of various parameters of seawater. Step S13: Determine the spectral changes of seawater based on the fiber optic whispering gallery mode according to the results of the changes of each parameter; analyze the spectral changes and mark the corresponding spectral shifts; Step S14: Determine multiple basic parameters by tracing along the spectral shift, and determine the comprehensive detection results of seawater under multiple parameters based on the multiple basic parameters.

[0013] refer to Figure 2In step S11, the specific steps are as follows: S111: Real-time monitoring of the multi-parameter detection process of seawater, and selection of corresponding theoretical analysis models of fiber optic microcavity whispering wall modes from the seawater detection database. In this theoretical analysis model, the mapping relationship between optical resonance conditions and cavity structure parameters is clarified. At this time, by analyzing the propagation characteristics of light at the interface between optically dense and optically sparse media, the critical incident angle, cavity wall refractive index difference, and cavity curvature radius that satisfy the formation of fiber optic microcavity whispering wall modes are determined, and the influence mechanism of geometric morphology on photon trapping ability is clarified. S112: Input the cavity structure parameters of the microbubble cavity of seawater, and set the dispersion characteristics, thermo-optic coefficient, and elastic modulus physical properties of fused silica. At the same time, set the corresponding matching layer boundary conditions according to the actual working conditions to construct a highly simulated microcavity geometric model. Input the microcavity geometric model into the optical-fluid coupling module. The optical-fluid coupling module analyzes the disturbance of the evanescent field in the cavity by the microfluidic liquid, and then performs multi-physics finite element analysis on the change of optical resonance characteristics of the microbubble cavity in the fluid environment to output multiple simulation results. The multiple simulation results reflect the interaction between the light field and the external environment.

[0014] S113: Collect multiple simulation results and perform multi-dimensional matching on the multiple simulation results to output different parameter combinations. Combine each parameter combination with the quality factor for multi-level iteration and determine the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor during the iteration process.

[0015] In the embodiments of this application, the multi-parameter detection process of seawater is monitored in real time, and the theoretical analysis model of the corresponding fiber optic microcavity whispering wall mode is selected from the seawater detection database. In the theoretical analysis model, the mapping relationship between the optical resonance condition and the cavity structure parameters is clarified. At this time, by analyzing the propagation characteristics of light at the interface between optically dense and optically sparse media, the critical incident angle, cavity wall refractive index difference and cavity curvature radius that satisfy the formation of the fiber optic microcavity whispering wall mode are determined, and the influence mechanism of geometric morphology on photon trapping ability is clarified.

[0016] At this time, the system acquires the range of physical parameters of the current monitored sea area in real time (such as the estimated temperature and salinity range); it searches in the pre-built "seawater detection database", which integrates mathematical and physical models of microbubble cavities with different geometric configurations (such as spherical, bottle-shaped, and capillary-shaped). These models are derived based on Maxwell's equations and describe the distribution law of the light field in the cavity; the screening process needs to consider the applicable boundaries of the model. For example, some models are specifically used for high refractive index liquid environments, while others are optimized for low-loss detection; the selected theoretical analysis model needs to be able to resolve the traveling wave transmission mode in the cavity and clarify the mathematical relationship between photon lifetime and cavity loss.

[0017] Optical resonance conditions are the cornerstone of the existence of whispering-gallery modes (WGMs). In this step, the abstract resonance equation needs to be transformed into a specific parameter mapping spectrum. Optical resonance conditions usually satisfy the phase matching principle. Through theoretical derivation, a functional mapping between cavity structural parameters (diameter D, wall thickness t, ellipticity e) and resonant wavelength offset and mode spacing is established. The focus is on analyzing the effect of structural micro-deformation on mode degeneracy relief and the influence of changes in cavity wall curvature radius on the optical field confinement potential well, thereby constructing a high-dimensional mapping matrix of "structural parameters-spectral response".

[0018] The formation of WGM relies on the principle of total internal reflection. This step delves into the microscopic physics level, quantitatively analyzing the behavior of light at the interface between quartz (optically denser medium, n≈1.45) and seawater / air (optically less dense medium). For the critical incident angle, Snell's law is used to derive the critical angle for total internal reflection, ensuring that the incident light angle is greater than this critical angle, thus forming a closed light path within the cavity wall. Regarding the refractive index difference of the cavity wall, the difference Δnn between the refractive index of the cavity wall material and the refractive index of the internal / external medium is determined. This difference determines the confinement strength of total internal reflection; the larger Δn is, the stronger the photon confinement ability and the lower the loss. Regarding the radius of curvature of the cavity, which directly affects the geometric optical trajectory of light, the diffraction loss of light after multiple reflections under a specific curvature needs to be calculated to determine the minimum radius of curvature required to maintain the WGM mode and prevent photon escape.

[0019] Ideally, a microbubble cavity should be a perfect sphere, but morphological defects exist in actual fabrication. Ellipticity influence: Analyze the degree to which the cavity cross-section deviates from a circle; ellipticity leads to mode splitting, disrupting degeneracy and thus affecting the Q-value (quality factor); the linear / nonlinear relationship between ellipticity and mode splitting distance needs to be clarified; Surface roughness: Analyze the scattering effect of nanoscale undulations in the cavity wall on photons; roughness is directly related to scattering loss and determines the theoretical limit Q-value; Wall thickness uniformity: Non-uniform wall thickness leads to local mode field volume changes, affecting the interaction strength between the evanescent field and the external environment. By establishing a correlation model between geometric morphology factor and photon trapping loss, the physical path of "morphological defects leading to photon escape" is revealed.

[0020] Specifically, in the seawater detection scenario, assuming the monitoring area is a near-shore estuary with large fluctuations in seawater salinity and a large amount of suspended matter, the system monitors in real time that the current environment is a "low salinity, high turbidity" condition. Based on this, the system selects the "anti-scattering interference type low-loss microbubble cavity theoretical model" from the database. This model has been specially modified for high scattering medium environment, which can effectively eliminate the background noise interference of seawater suspended particles on the light field distribution, and provide an accurate theoretical framework for subsequent calculations.

[0021] Since changes in salinity cause slight fluctuations in refractive index, the model needs to focus on outputting the mapping relationship between "wall thickness and refractive index sensitivity". For example, the theoretical model calculates that when the wall thickness of the microbubble cavity is controlled at about 1 micrometer, the evanescent field penetration depth is most suitable for detecting changes in the refractive index of seawater. At this time, the mapping relationship clearly shows that for every 10 micrometers increase in the cavity diameter, the free spectral range (FSR) will shift by a specific value, thereby guiding the size selection of the subsequent sensor design and ensuring that the resonance peak will not change mode due to the temperature drift of seawater.

[0022] Seawater, as an optically less dense medium, fills the interior of the microbubble cavity. Assuming the refractive index of seawater is nseawater≈1.33 and the refractive index of the quartz cavity wall is nsilica≈1.45, the system calculates the critical angle for total internal reflection to be 66.3 degrees. This means that the propagation angle of light within the cavity must be greater than this value to form a whispering gallery mode. Simultaneously, considering the temperature stratification phenomenon in seawater, the system dynamically calculates the refractive index difference under different temperature gradients to ensure that Δn still satisfies the total internal reflection condition even at extreme temperatures. Furthermore, to reduce scattering losses caused by impurities in the seawater, the determined radius of curvature must be combined with specific focusing parameters to maximize the photon trapping efficiency.

[0023] The sensor is required to have an extremely high Q value to capture minute changes in salinity. System analysis revealed that the slight ellipticity of the microcavity causes bimodal splitting in the WGM spectrum, which can easily lead to misinterpretation in the complex spectral background of seawater. Therefore, this step clarified the mechanism that "for every 0.1% increase in ellipticity, mode splitting intensifies by X times," thus guiding the introduction of a morphology correction factor when building the model in the subsequent S112 step. For example, for seawater, the morphology parameters with the strongest photon trapping ability were determined to be: ellipticity error <0.5% and surface roughness <10nm, which were used as the theoretical boundary conditions for microcavity fabrication.

[0024] Furthermore, the cavity structure parameters of the seawater microbubble cavity are input, and the dispersion characteristics, thermo-optic coefficient, and elastic modulus physical properties of fused silica are set. At the same time, the corresponding matching layer boundary conditions are set according to the actual working conditions to construct a highly simulated microcavity geometric model. This microcavity geometric model is input into the optical-fluid coupling module, which analyzes the disturbance of the microfluidic liquid on the evanescent field inside the cavity. Then, multi-physics finite element analysis is performed on the changes in the optical resonance characteristics of the microbubble cavity in the fluid environment to output multiple simulation results. The multiple simulation results reflect the interaction between the light field and the external environment.

[0025] At this point, input the optimized cavity geometry parameters from step S111, including the outer diameter of the microbubble cavity (e.g., 100-500 μm), wall thickness (target <2 μm), and cavity ellipticity. In the material property settings, precise values ​​need to be assigned based on the physical properties of fused silica: Dispersion characteristics: The Sellmeier equation is used to define the nonlinear relationship between the refractive index of silica and wavelength to ensure the accuracy of the transmission of the simulated light field at different wavelengths; The thermo-optic coefficient is used to characterize the change in the refractive index of the material caused by temperature changes, which is the physical basis of temperature sensing; Elastic modulus and Poisson's ratio: Set the Young's modulus (approximately 73 GPa) and Poisson's ratio of silica to calculate the mechanical deformation of the cavity wall under fluid pressure. Through the above settings, the response benchmark of the microcavity under the action of multiple physics fields is established.

[0026] The constructed geometric model is imported into the optical-fluid coupling module (such as the electromagnetic wave and fluid flow interface in COMSOL Multiphysics). Most of the light field energy in the whispering gallery mode is confined within the cavity, but an exponentially decaying evanescent field exists on the cavity wall surface, extending into the fluid within the cavity. When seawater flows through the microbubble cavity, changes in its velocity and viscosity coefficient cause mechanical vibrations (Brownian motion or forced vibration) in the cavity wall. Simultaneously, changes in the refractive index of the seawater directly alter the effective refractive index of the evanescent field. The optical-fluid coupling module calculates the perturbation of the evanescent field distribution caused by changes in fluid parameters by solving Maxwell's equations and fluid dynamics equations.

[0027] The finite element method is used to solve the multiphysics coupling equations and output the key optical characteristic parameters of the microbubble cavity in a fluid environment; the resonant wavelength drift of different modes (fundamental mode and higher-order radial mode) is calculated; the Q value change in the fluid environment is calculated by combining radiation loss, material absorption loss and fluid viscous loss; the changes in transmission spectrum or reflection spectrum are output, and the displacement (wavelength drift) and broadening (linewidth change) of resonance peaks are marked.

[0028] Specifically, in order to simulate an infinitely large external environment and absorb outgoing waves in finite element analysis (FEA), a matching layer boundary condition needs to be set. Perfectly matched layer (PML): A PML is set at the boundary of the air or fluid domain around the microcavity to fully absorb the outwardly radiated light field and prevent boundary reflection from interfering with the resonance characteristics of the whispering gallery mode, thereby accurately calculating radiation loss. For fluid-solid coupling boundary, fluid-structure interaction (FSI) boundary conditions are set on the inner surface of the cavity wall in contact with seawater to ensure the interaction between the fluid dynamics equations (Navier-Stokes equations) and the solid mechanics equations, simulating the mechanical loading effect of fluid pressure on the cavity wall.

[0029] Seawater serves as the fluid medium, filling the interior of the microbubble cavity. The model sets the internal medium properties to seawater (refractive index approximately 1.33-1.34) and sets PML boundaries on the outer wall of the cavity and the tapered coupling region of the optical fiber to simulate the open environment of the deep sea. Meanwhile, considering the scattering loss caused by suspended particles in the seawater, a weak scattering loss factor is introduced into the boundary conditions to construct a "high-fidelity microcavity geometric model" that closely approximates the real marine environment.

[0030] Because changes in salinity cause slight fluctuations in the refractive index (for every 1 unit increase in salinity, the refractive index increases by approximately 1.5 × 10⁻⁶),... - 4 (RIU) The model needs to focus on analyzing the response of the evanescent field to changes in refractive index; in the optical-fluid coupling module, the process of seawater salinity changing from 30 to 35 is simulated, and the shift of the energy distribution of the evanescent field in the cavity is calculated; the module quantifies the dependence of the evanescent field penetration depth on the refractive index of seawater, reveals the degree of leakage of optical field energy into the fluid, and thus establishes the physical relationship between "optical field disturbance - seawater refractive index change".

[0031] Based on the characteristics of seawater, the model focuses on outputting the mapping relationship between "wall thickness and refractive index sensitivity". Through multiple simulation comparisons, it was found that when the microbubble cavity wall thickness is set to 1.8 μm, the simulation results show that the refractive index sensitivity of seawater is approximately 800 nm / RIU; when the wall thickness is optimized to 1.0 μm, the sensitivity increases to over 1200 nm / RIU due to the significant increase in the proportion of evanescent field penetrating the cavity wall. The simulation results are output in the form of a data matrix, clearly demonstrating the nonlinear positive correlation between microcavity wall thickness and refractive index detection sensitivity. Based on this, researchers can determine the optimal wall thickness parameter for seawater detection (e.g., choosing 1.2 μm as the best value to balance sensitivity and mechanical stability).

[0032] Therefore, multiple simulation results were collected and multi-dimensionally matched to output different parameter combinations. Each parameter combination was combined with the quality factor for multi-level iteration. During the iteration process, the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor was determined, thus introducing the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor.

[0033] At this point, the system collects a dataset of simulation results including light field distribution, resonant frequency, radiation loss, and mechanical vibration modes. Using a multi-dimensional matching method, the spectral features of the simulation output (such as resonance wavelength, linewidth, and free spectral range FSR) are compared with a preset theoretical feature library. The matching dimensions cover: geometric dimension: mode distribution under different cavity diameter and wall thickness combinations; physical dimension: stress tensor distribution under different flow rates and pressures; optical dimension: effective refractive index variation under different environmental refractive indices. Through dimensionality reduction and cluster analysis, invalid or distorted simulation points are eliminated, and a representative, physically and logically consistent set of "structure-performance" parameter combinations is output.

[0034] The selected parameter combinations are input into the iterative optimization method, with the primary objective of maximizing the quality factor Q. Sensitivity is introduced as a constraint. In multi-level iterations, the cavity diameter is adjusted to determine the approximate free spectral range and avoid mode overlap. The wall thickness and curvature are finely adjusted to optimize the light field constraint factor. Each iteration recalculates the light field distribution and loss mechanism (scattering loss, absorption loss, radiation loss) and feeds back to correct the cavity parameters until the Q value converges to the theoretically expected value.

[0035] After multiple iterative calculations, the system finally establishes a quantitative functional relationship or mapping spectrum between cavity structural parameters (independent variables) and quality factor Q (dependent variable). This relationship typically exhibits nonlinear physical laws: wall thickness influence: establishing the critical threshold at which thinning of the wall leads to a decrease in the radial mode number but an increase in radiation loss; surface roughness influence: quantifying the limiting effect of Rayleigh scattering caused by nanoscale roughness on the Q value; ellipticity influence: clarifying the mathematical relationship between mode splitting caused by deformation and the broadening of the Q value. Finally, a deterministic design rule base or sensitivity matrix is ​​output to guide the subsequent physical fabrication process of microbubble cavities.

[0036] Specifically, step S112 outputs hundreds of sets of spectral data under different salinity (e.g., 30 to 35) and temperature gradients. After acquiring these results, the system focuses on matching the fundamental mode features in the spectrum to eliminate interference from higher-order radial modes. Considering the small fluctuations in seawater refractive index, the system extracts discrete data pairs of "wall thickness (e.g., 0.8 μm, 1.2 μm, 1.5 μm)" and "wavelength shift". Through multi-dimensional matching, a specific set of parameter combinations is selected: when the wall thickness is less than 1.5 μm, the spectral shift shows the best linearity with salinity changes, and the Q value does not decrease sharply due to the thinner cavity wall.

[0037] During the iteration process, it was found that as the wall thickness decreased, the evanescent field penetration depth increased, and the salinity sensitivity improved, but the Q value decreased due to increased radiation loss. The system underwent multiple iterations to weigh the trade-offs: In the first iteration, the outer diameter was fixed at 300 μm, and it was found that the Q value was highest when the wall thickness was 1.0 μm, but the sensitivity was only moderate. In the second iteration, the signal-to-noise ratio (SNR) was introduced as a weight, and the wall thickness was fine-tuned to 1.1 μm. At this point, although the Q value decreased slightly, the SNR for salinity detection reached the optimal level. The iteration process successfully balanced the contradiction between "high sensitivity" and "low loss" in seawater detection.

[0038] Based on the characteristics of seawater, the system ultimately outputs a mapping spectrum of "wall thickness-refractive index sensitivity". The results show that when the wall thickness is >2.0 μm, the evanescent field is mainly confined within the quartz wall, resulting in low sensitivity to changes in the refractive index of seawater (<50 nm / RIU), but an extremely high Q value. When the wall thickness is ≈1.0 μm, the evanescent field significantly penetrates into the seawater, and the sensitivity jumps to >1000 nm / RIU, while the Q value remains at the corresponding order of magnitude. When the wall thickness is <0.5 μm, although the sensitivity is high, the Q value drops sharply due to photon leakage, making it impossible to maintain resonance. Based on this, the optimal cavity structure parameters for seawater are determined to be: a wall thickness of 1.0 ± 0.2 μm. This parameter combination can accurately capture minute fluctuations in the refractive index caused by changes in seawater salinity while ensuring a high quality factor.

[0039] refer to Figure 3 In step S12, the specific steps are as follows: S121: Collect the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor, analyze the influence relationship, and identify multiple influence nodes during the analysis process. Mark the control mechanism of the corresponding parameter of each influence node; perform corresponding finite element analysis on multiple influence nodes in different dimensions, and gradually construct a multiphysics coupling model of the microbubble cavity during the analysis process. S122: A multiphysics coupling model for real-time monitoring of microbubble cavities. This model covers the resonant characteristics of the corresponding fiber whispering gallery mode and, by introducing the optical-mechanical-fluid coupling effect, further simulates the physical process of seawater flowing and filling the microbubble cavity as a microfluidic liquid. Combined with the heat transfer module and stress analysis module, it simulates the temperature gradient field distribution and stress distribution inside and outside the cavity wall generated when seawater is transported in the microbubble cavity. It quantitatively determines the influence of the effective refractive index of the microcavity and the cavity geometry, and marks the changes in various parameters of the seawater, including the temperature, pressure and flow velocity of the seawater.

[0040] In the embodiments of this application, the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor is collected, the influence relationship is analyzed, and multiple influence nodes are identified during the analysis process. The control mechanism of the corresponding parameter is marked for each influence node. Finite element analysis is performed on the multiple influence nodes in different dimensions, and a multiphysics coupling model of the microbubble cavity is gradually constructed during the analysis process, which improves the accuracy of the multiphysics coupling model of the microbubble cavity.

[0041] At this point, the established "structural parameter-Q value" mapping relationship is imported into the data processing module. The analysis process is no longer limited to simple parameter correspondence, but delves into the underlying physical loss mechanism. The system decomposes the total loss, which is the reciprocal of the Q value, into multiple components: material absorption loss, radiation loss, scattering loss, and fluid viscous loss. By analyzing the slope and inflection point of the influence relationship curve, the dominant factors restricting the improvement of the Q value are identified. For example, when the wall thickness is reduced to a certain threshold, the radiation loss increases sharply, and this threshold is the key feature point for analysis.

[0042] Based on the analysis of the influence relationships, the system automatically identifies key "influence nodes," which represent physical state abrupt changes or sensitivity extremes under specific structural parameters; it identifies points such as "critical coupling points," "maximum sensitivity points," and "mechanical resonance intersections"; and it assigns a physical label to each node; for example, for wall thickness parameters, its regulation mechanism is labeled as "evanescent field penetration depth control"; for cavity curvature, it is labeled as "photon trapping potential well modulation"; each node corresponds to a specific physical equation (such as the boundary conditions of Maxwell's equations and the Navier-Stokes equations of fluid dynamics), clarifying how adjusting this node can affect the final performance of the sensor.

[0043] For the identified influencing nodes, coupling analysis is performed in multiple dimensions (optics, mechanics, thermodynamics, and fluid dynamics): Optical-thermal coupling dimension: analyzing the impact of thermal expansion and thermo-optical effects caused by temperature gradients on the resonant frequency drift; Fluid-structure interaction (FSI) dimension: analyzing the mechanical deformation of the cavity wall caused by fluid pressure, and the impact of deformation on the optical path length; Optical-mechanical-fluid coupling dimension: focusing on simulating the damping effect of fluid viscous drag on the mechanical vibration mode of the microcavity, and its broadening effect on the optical resonant spectral linewidth. By applying multiphysics loads at each node, a complete multiphysics coupling model of the microcavity, including optical field distribution, stress-strain distribution, and flow field distribution, is gradually established.

[0044] Specifically, the system collected the "wall thickness-Q value" dataset output from step S113. Analysis revealed that when the wall thickness is less than 1.0 μm, the quality factor drops sharply. Further analysis of the loss mechanism indicates that this is due to the excessive leakage of evanescent field energy to the external environment (air or seawater) caused by the excessively thin cavity wall, resulting in a surge in radiation loss. At the same time, considering the high salinity of seawater, the analysis also revealed the sensitivity of material absorption loss to salinity changes at specific wavelengths, providing a physical basis for subsequent node labeling.

[0045] For seawater salinity monitoring, the system identified a key influencing node at a wall thickness of 1.2 μm—the "sensitivity saturation node," which was labeled as the "evanescent field-fluid refractive index coupling node." The system recorded the regulation mechanism of this node as "adjusting the interaction strength between the energy density inside the cavity and the refractive index of the external fluid through wall thickness." Specifically, when the wall thickness is less than this node value, although the evanescent field is enhanced, the Q value decreases, leading to a reduction in the signal-to-noise ratio. When the wall thickness is greater than this node value, the evanescent field is too weak to effectively detect the minute refractive index fluctuations of seawater. The establishment of this node provides precise boundary constraints for constructing a multiphysics model.

[0046] The system focuses on finite element analysis of the mapping relationship between wall thickness and refractive index sensitivity. At the sensitivity saturation node (wall thickness 1.2 μm), a coupled model is constructed that includes a seawater fluid domain, a quartz cavity wall solid domain, and a light wave propagation domain. Multi-dimensional analysis includes: Optical dimension: calculating the penetration depth of the evanescent field in seawater at this wall thickness to confirm its sufficient coverage of the sensitive layer for salinity changes; Fluid-mechanical dimension: simulating the microflow state of seawater within the microbubble cavity and calculating the mechanical disturbance of the thin wall (1.2 μm) by fluid shear stress to ensure that the cavity does not experience structural instability in a fluid environment; Coupled output: the model ultimately outputs the drift of the microbubble cavity resonant wavelength when seawater salinity changes (e.g., from 30 to 35) cause small fluctuations in refractive index. The results verify that at the influence node with a wall thickness of 1.2 μm, the model can accurately predict the refractive index sensitivity and successfully decouple the cross-sensitivity effect caused by temperature changes, thus completing the technical closed loop from single structural parameter optimization to multi-physics integrated modeling.

[0047] Furthermore, a multiphysics coupling model of the microbubble cavity is monitored in real time. This model encompasses the resonant characteristics of the corresponding fiber whispering gallery mode. By introducing the optical-mechanical-fluid coupling effect, the physical process of seawater flowing and filling the microbubble cavity as a microfluidic liquid is further simulated. Combining the heat transfer module and the stress analysis module, the temperature gradient field distribution and stress distribution inside and outside the cavity wall generated during seawater transmission within the microbubble cavity are simulated. The influence of the effective refractive index of the microbubble cavity and the cavity geometry is quantitatively determined, and the results of various parameter changes of seawater, including temperature, pressure, and flow velocity, are marked. At the same time, the influence of the cavity structure parameters of the microbubble cavity on the quality factor is introduced, and the multiphysics coupling model of the microbubble cavity is further controlled, improving the simulation effect of the temperature gradient field and stress distribution, and further improving the accuracy of the results of various parameter changes.

[0048] At this point, the light field distribution is solved based on Maxwell's equations, while the Navier-Stokes equations are coupled to describe the fluid motion. The core lies in introducing a "optic-mechanical-fluid" coupling mechanism: Optical-fluid coupling: the perturbation of the evanescent field in the microcavity by the microfluidic flow is calculated, and the fluid velocity distribution affects the effective refractive index distribution of the light field; Mechanical-fluid coupling: the fluid dynamic pressure acts on the inner wall of the microcavity, causing micro-deformation of the cavity, which in turn changes the geometric perimeter of the resonant cavity; Process simulation: the entire process of seawater from injection, filling to stable flow is simulated, and the transient effects of different fluid states (laminar flow, turbulent flow) on the stability of the WGM mode are observed.

[0049] Multiphysics simulation was performed using the heat transfer and solid mechanics modules of the finite element method (FEM) software: Heat transfer simulation: Solving the heat conduction equation to simulate the heat exchange between seawater temperature and the quartz wall of the microcavity; focusing on analyzing the temperature gradient field inside the cavity caused by changes in seawater temperature, and identifying the thermal equilibrium time constant; Stress analysis simulation: Calculating the stress tensor distribution generated on the thin wall of the microcavity by the pressure difference between external hydraulic pressure (depth pressure) and internal fluid pressure, and using the coefficient of thermal expansion to couple the calculation of the superposition effect of thermal stress and mechanical stress; outputting temperature cloud maps and stress concentration areas inside and outside the cavity wall to identify weak points that are prone to fracture or sudden changes in optical performance.

[0050] The key to transforming physical field simulation results into optical parametric quantization outputs is to use thermo-optical and elasto-optical coefficients to map the simulation results of temperature and stress fields to changes in the refractive index of the microcavity material; combined with the fluid refractive index (affected by temperature and salinity), the effective refractive index of the WGM mode is comprehensively calculated; at the same time, the changes in cavity diameter caused by temperature and pressure are calculated using the coefficient of thermal expansion and elastic modulus; and based on the resonance condition, the amount of resonant wavelength drift caused by changes in effective refractive index and RR is quantitatively calculated.

[0051] Specifically, the system's real-time monitoring model simulates the microfluidic process of seawater within the microbubble cavity. Since seawater contains suspended particles, the model focuses on simulating the weak shear stress generated by the fluid dynamics viscous resistance on the thin wall (1.2 μm thick) of the microcavity. Monitoring revealed that when seawater flows through the cavity at a specific velocity, the fluid pressure causes the cavity to undergo nanoscale radial expansion. This mechanical deformation directly modulates the resonant frequency of the WGM. The model successfully captured the "additional optical path effect" caused by fluid flow, namely, the change in optical path difference caused by changes in flow velocity.

[0052] For actual seawater conditions (such as temperature stratification), the model simulated the heat transfer process when the seawater temperature drops sharply from 25°C to 15°C. The results show that the microbubble cavity wall is extremely thin, the heat conduction rate is extremely fast, and the temperature gradient field reaches equilibrium within milliseconds. At the same time, the stress analysis module calculated the cavity wall stress distribution at different depths in seawater (corresponding to different hydrostatic pressures). The simulation found that the temperature decrease causes the quartz material to shrink (thermal stress), while the increase in external water pressure causes the cavity wall to be compressed (mechanical stress). The superposition effect of the two changes the effective perimeter of the cavity, which provides key physical parameters for the subsequent decoupling of temperature and depth signals.

[0053] Based on the characteristics of seawater, the model focuses on outputting the contribution of the "wall thickness-refractive index sensitivity" mapping relationship to the spectral response; when the seawater salinity increases and the refractive index increases, the model quantitatively calculates that the effective refractive index increases accordingly; the results show that at the optimization node with a wall thickness of 1.2 μm, the sensitivity of the evanescent field to the seawater refractive index reaches its peak (approximately 1200 nm / RIU).

[0054] The system marks the specific parameter changes: a 1°C change in temperature causes a wavelength shift of approximately 15 pm (mainly due to the thermo-optical effect); a 1 MPa change in pressure causes a wavelength shift of approximately 100 pm (mainly due to the elasto-optical effect and cavity deformation); and the refractive index fluctuations caused by salinity changes are directly converted into wavelength shifts through the "wall thickness-refractive index sensitivity" mapping relationship, thereby achieving accurate marking and decoupling of seawater temperature, pressure, and flow velocity parameters.

[0055] refer to Figure 4 In step S13, the specific steps are as follows: S131: Collect the results of various parameter changes, and determine the intrinsic mapping mechanism between seawater parameter changes and spectral drift by combining the characteristics of spectral drift. In this mapping mechanism, analyze the generated transmission spectrum data, focus on monitoring the dynamic evolution of the center wavelength position, peak intensity and spectral linewidth of the echo-gallery mode resonance peak, identify the spectral feature change patterns caused by different seawater parameters, capture the corresponding spectral change content, and present the characteristics of spectral morphology changing with the seawater environment. S132: In this spectral change content, the spectrum after environmental disturbance is marked, the reference spectrum is compared with the spectrum after environmental disturbance, the drift of the resonant wavelength is extracted, the drift of the resonant wavelength is used as the core sensing signal, and the spectral response differences of different order modes are distinguished to be converted into quantifiable spectral shift, and a quantitative correspondence between spectral shift and changes in multiple parameters of seawater is established.

[0056] In the embodiments of this application, the results of various parameter changes are collected, and the intrinsic mapping mechanism between seawater parameter changes and spectral drift is determined by combining the characteristics of spectral drift. In this mapping mechanism, the generated transmission spectrum data is analyzed, and the dynamic evolution of the center wavelength position, peak intensity and spectral linewidth of the echo-gallery mode resonance peak is monitored. The spectral feature change patterns caused by different seawater parameters are identified, the corresponding spectral change content is captured, and the characteristics of spectral morphology changing with the seawater environment are presented.

[0057] At this point, the system collects the temperature, pressure, flow rate, and refractive index change data output from step S122 and uses them as input variables; simultaneously, it collects the corresponding simulated transmission spectrum data and extracts spectral drift characteristics (such as wavelength shift) as output variables; based on the resonance condition, it analyzes the independent contribution of each parameter to the resonant wavelength: temperature mapping: establishing the temperature-wavelength drift relationship through thermo-optical effect and thermal expansion effect; pressure mapping: establishing the pressure-wavelength drift relationship through elastic-optical effect and mechanical deformation; refractive index mapping: using the evanescent field principle, establishing the coupling relationship between the external refractive index (salinity) and the effective refractive index; the system finally outputs a multi-dimensional "parameter-spectrum" mapping matrix, decoupling the contribution weights of each parameter to the spectral drift.

[0058] High-precision analysis of the simulated transmission spectrum was performed using spectral analysis to extract three key feature parameters: center wavelength position: the resonant peak and valley values ​​were accurately located using Lorentz fitting, achieving a resolution at the picometer level, directly reflecting changes in effective refractive index and cavity size; peak intensity: the extinction ratio of the transmission spectrum was monitored, reflecting the coupling efficiency and loss of the light field within the cavity; increased fluid absorption or scattering loss leads to a decrease in peak intensity; spectral linewidth: the full width at half maximum (FWHM) of the resonant peak was calculated, directly correlated with the quality factor Q; linewidth broadening usually indicates increased scattering loss (e.g., suspended particles) or mode splitting. By dynamically monitoring these three parameters, the real-time evolution trajectory of the spectral characteristics was constructed.

[0059] The system uses machine learning or feature thresholding to match the extracted spectral parameter changes with known physical mechanisms, identifying specific "spectral information": single parameter patterns, such as temperature changes alone manifesting as an overall translation with essentially unchanged linewidth; coupled parameter patterns, such as temperature-refractive index dual-parameter changes, manifesting as a specific combination of wavelength drift and linewidth broadening; and anomalous interference patterns, such as spectral distortion and splitting, which usually correspond to mechanical vibration or strong scattering interference. Finally, these spectral changes are captured and labeled to form a visualized spectral morphology evolution map, providing solid data support for the subsequent parameter inversion in step S14.

[0060] Specifically, the system collected two sets of data: one for increased temperature (e.g., 25°C to 30°C) and the other for increased salinity (e.g., 30°C to 35°C). The mapping mechanism analysis revealed that the spectral "redshift" (lengthening of wavelength) caused by increased temperature was mainly dominated by the thermo-optical effect of quartz. The slight fluctuations in refractive index caused by increased salinity, through the "wall thickness-refractive index sensitivity" mapping relationship determined in step S11, also caused a spectral "redshift". To distinguish between the two, the system established a cross-sensitive mapping mechanism: it was found that temperature changes cause a slight broadening of the spectral linewidth (increased thermal noise), while salinity changes mainly affect the extinction ratio of the resonance peak (ModeContrast). Through this multidimensional mapping, the system initially distinguishes the spectral drift characteristics caused by different parameters.

[0061] The system analyzes the simulated transmission spectrum frame by frame; center wavelength: the wavelength was detected to drift from 1550nm to 1550.150nm, and the system recorded this drift; peak intensity: it was found that the peak intensity slightly decreased with the increase of seawater flow velocity, which is due to optical coupling jitter caused by hydrodynamic instability; spectral linewidth: in view of the risk of microbial attachment in seawater, the system focuses on monitoring the evolution of linewidth; the simulation shows that if the viscosity coefficient of seawater increases, it will lead to an increase in mechanical damping, which in turn causes a slight broadening of the spectral linewidth, and the dynamic process of the linewidth evolving from 0.1pm to 0.15pm was successfully captured.

[0062] Based on the characteristics of seawater, the system focused on identifying spectral feature modes related to "wall thickness-refractive index sensitivity". At the wall thickness node optimized in step S113 (e.g., 1.2 μm), the system found that the spectral drift mode caused by salinity changes has a high signal-to-noise ratio. When the seawater salinity fluctuates slightly (e.g., 0.1‰), the system detected a shift of about 12 pm in the center wavelength, and the spectral morphology remained good (without severe distortion).

[0063] The system outputs characteristic spectra of spectral morphology changes with the environment, clearly showing the composite mode in the seawater environment where the resonance peak smoothly redshifts with increasing salinity and further redshifts with increasing temperature. By comparing the "wall thickness-refractive index sensitivity" mapping relationship, the system confirms that at this wall thickness, the wavelength shift caused by salinity is significantly greater than the conventional noise level, thus successfully identifying the key characteristic mode of salinity change.

[0064] Furthermore, in this spectral change content, the spectrum after environmental disturbance is marked, the reference spectrum and the spectrum after environmental disturbance are compared, the drift of the resonant wavelength is extracted, the drift of the resonant wavelength is used as the core sensing signal, and the spectral response differences of different order modes are distinguished to be converted into quantifiable spectral shift, and a quantitative correspondence between spectral shift and changes in multiple parameters of seawater is established.

[0065] At this point, the system establishes a "reference spectrum" under zero-disturbance conditions, namely the transmission spectrum of the microbubble cavity under standard temperature (e.g., 25℃) and pure water environment; the system collects the "dynamic spectrum" after environmental disturbance in real time, and uses cross-correlation function or spectrum matching method to align and compare the dynamic spectrum with the reference spectrum in the wavelength domain; the marked content includes: changes in the position of transmission valley, distortion of resonance peak shape and fluctuation of background noise. Through differential spectroscopy technology, the spectral differences caused by environmental disturbance are presented intuitively, eliminating the inherent baseline drift of the system.

[0066] Using Lorentz fitting or Gaussian fitting, the resonance valleys in the reference spectrum and the perturbation spectrum are located at the subpixel level with an accuracy of femtometer. Since the whispering gallery mode is extremely sensitive to environmental parameters (refractive index, temperature, stress), the wavelength shift directly characterizes the changes in the effective refractive index and cavity radius, and therefore it has been established as the core sensing signal.

[0067] Different orders of radial modes (fundamental mode vs. higher-order modes) have different optical field distributions, resulting in different sensitivities to environmental parameters. The system needs to identify and distinguish the responses of these modes. For example, the fundamental mode usually has the highest Q value and is suitable for high-precision detection, while higher-order modes have different sensitivity coefficients to changes in the external refractive index.

[0068] The key to inverting optical signals into physical parameters is to establish a multi-parameter sensing matrix equation; convert the extracted wavelength drift into a dimensionless or standard-unit "spectral shift"; use the multi-physics coupling model coefficients (temperature coefficient, refractive index sensitivity, pressure sensitivity) obtained in step S12 to construct a set of equations; by monitoring the differences in temperature and refractive index sensitivity of different order modes, decouple the cross-sensitivity effect, thereby establishing a precise one-to-one correspondence and realizing multi-parameter synchronous inversion.

[0069] Specifically, in seawater detection, the reference spectrum is set as the steady-state transmission spectrum at a salinity of 35‰ and a temperature of 25℃. When the seawater salinity fluctuates slightly (such as dropping to 34.5‰) or the temperature changes, the system collects the real-time spectrum and superimposes it with the reference spectrum. The comparison reveals that the dynamic spectrum has shifted overall, and some higher-order modes have shown slight amplitude changes. The system accurately marks these disturbed areas, eliminates random scattering noise caused by suspended particles in the seawater, and locks in the effective sensing signal area.

[0070] The system focuses on monitoring the fundamental mode (m-order) response in a microbubble cavity with a wall thickness of 1.2 μm. It detects that the decrease in seawater salinity leads to a decrease in refractive index, which in turn causes the resonant wavelength to drift towards shorter wavelengths. The drift amount is extracted as -5.2 pm. Analysis shows that the fundamental mode has the most linear response to this refractive index change, while a certain higher-order radial mode, although more sensitive, is more affected by the fluctuations caused by seawater flow velocity. Therefore, the system locks the fundamental mode wavelength drift amount as the core sensing signal and, combined with the "wall thickness-refractive index sensitivity mapping" relationship, confirms the refractive index change corresponding to this drift amount, thereby eliminating the measurement uncertainty caused by higher-order modes.

[0071] The system establishes the following quantitative relationship: Assuming the total spectral shift detected is +15 pm, the temperature rise is measured to be 0.5℃ using a reference optical fiber (not in contact with seawater, only sensing temperature). Combining this with a temperature coefficient of 10 pm / ℃, the temperature-induced shift is calculated to be +5 pm. The depth sensor measures no change in pressure, so the pressure term is 0. The remaining +10 pm shift is attributed to the change in seawater refractive index (salinity). Based on the "wall thickness-refractive index sensitivity" mapping relationship output from step S11 (assuming a sensitivity of 1000 nm / RIU), the system determines that the seawater refractive index has increased. Thus, the system successfully establishes a quantitative correspondence between "spectral shift +15 pm" and "temperature +0.5℃, salinity +0.1‰", achieving decoupling and accurate detection of multiple parameters.

[0072] refer to Figure 5 In step S14, the specific steps are as follows: S141: The spectral shift is traced back, and the corresponding spectral shift differences are determined during the tracing process. The matrix inversion is used to determine each basic parameter, and the independent contribution of each basic parameter to the spectral shift is marked. Deep learning is performed on each basic parameter, and cross-sensitive interference factors in the measurement of a single parameter are eliminated during the learning process. Each basic parameter covers the temperature, salinity, density, pressure and flow velocity of seawater. S142: Data fusion of various basic parameters is performed, and different levels of sub-detection results are divided in sequence during the fusion process. Based on the multiple sub-detection results at different levels, the comprehensive detection result of seawater under multiple parameters is determined. This comprehensive detection result not only includes the value of a single physical quantity, but also draws a dynamic distribution map of seawater through multi-dimensional cross-validation.

[0073] In the embodiments of this application, the spectral shift is traced back, and the corresponding spectral shift difference is determined during the tracing process. The matrix inversion is combined to determine each basic parameter, and the independent contribution of each basic parameter to the spectral shift is marked. Deep learning is performed on each basic parameter, and cross-sensitive interference elements in the measurement of a single parameter are eliminated during the learning process. Each basic parameter covers the temperature, salinity, density, pressure and flow velocity of seawater.

[0074] At this point, by tracing the source from the total spectral shift signal, the system analyzes the differences in spectral characteristics caused by different physical mechanisms. The system not only focuses on the wavelength shift amount but also traces the differences in spectral morphology. The tracing process utilizes multi-parameter decoupling theory: Wavelength shift differences: Temperature and salinity changes usually cause an overall shift of the resonance peak, and the directions are the same (e.g., both cause redshift), so it is necessary to trace their minute differences; Linewidth and intensity differences: Pressure (depth) changes cause cavity deformation, leading to mode splitting or linewidth broadening, and flow velocity changes cause intensity modulation. By comparing with a standard simulated spectral library, the system identifies whether the current spectral shift is dominated by a single parameter or by multi-parameter coupling, and quantifies its "spectral shift difference" feature vector.

[0075] According to sensing theory, the spectral shift is a linear superposition of the changes in the fundamental parameters (within a small range). By establishing a sensing matrix equation and monitoring multiple modes (such as the fundamental mode and the first-order radial mode) or multiple spectral features (wavelength and linewidth), a system of equations is constructed. The fundamental parameter change vectors (ΔT, ΔS, ΔP) are solved using matrix inversion operations. The independent contribution of each parameter is marked in the system.

[0076] To address the errors and nonlinear cross-sensitivity issues inherent in linear matrix inversion under complex environments, the system incorporates deep learning methods (such as backpropagation neural networks or convolutional neural networks, CNNs). Using spectral features (wavelength, intensity, linewidth, morphological fingerprints) as the input layer and basic seawater parameters (temperature, salinity, density, pressure, flow velocity) as the output layer, the network is trained using extensive experimental and simulated data to learn the nonlinear coupling relationships between parameters. During training, the weights of cross-sensitivity terms are identified and reduced; for example, the interference of temperature on refractive index measurement is automatically compensated. The network outputs purified parameter values ​​after "denoising" and "decoupling," thus achieving high-precision detection.

[0077] Specifically, in seawater monitoring, the system detected a total spectral shift of +15 pm. Tracing back to the source revealed two main components: a +15 pm shift in the main resonance peak; and an increase of 0.02 pm in the full width at half maximum (FWHM). The spectral shift differences determined by the system indicate that the shift primarily corresponds to changes in refractive index or temperature, while the linewidth broadening corresponds to changes in the distribution of mechanical stress caused by pressure. This difference provides a basis for subsequently distinguishing salinity from depth signals.

[0078] Based on the "wall thickness-refractive index sensitivity" mapping relationship determined in step S11 (e.g., sensitivity is 1200 nm / RIU when the wall thickness is 1.2 μm), the system constructs a decoupling matrix: temperature sensitivity KT≈10 pm / ℃; salinity sensitivity KS≈1 pm / 0.1‰ (based on refractive index mapping conversion); pressure sensitivity KP≈0.5 pm / m; substituting the total shift +15 pm and combining it with the linewidth variation data, the matrix inversion calculation yields: temperature contribution: +5 pm (corresponding to ΔT=+0.5℃); salinity contribution: +9 pm (corresponding to ΔS=+0.9‰); pressure contribution: +1 pm (corresponding to ΔP≈2m); the system successfully marks the independent contributions of each basic parameter, accurately distinguishing the small salinity fluctuations caused by freshwater injection from the ambient temperature drift.

[0079] Temperature fluctuations are highly correlated with salinity changes (e.g., solar radiation causes temperature increases while evaporation causes salinity increases); the system inputs continuously monitored spectral sequences into a deep learning model; the model identifies that at a specific wall thickness, increased temperature leads to an increase in the refractive index of quartz (spectral redshift), and increased salinity also leads to a spectral redshift.

[0080] By learning from historical data, the model discovered that temperature changes are accompanied by subtle fluctuations in spectral linewidth, while salinity changes are not. Utilizing this implicit feature, deep learning effectively eliminates the cross-sensitivity interference of temperature on salinity measurements. Even during the day and night when temperatures fluctuate drastically, the system's output seawater salinity data remains stable, accurately reflecting the true subtle fluctuations in salinity, and achieving comprehensive high-precision detection of temperature, salinity, density, pressure, and flow velocity.

[0081] Furthermore, the data of each basic parameter is fused, and the sub-detection results at different levels are divided in sequence during the fusion process. Based on the multiple sub-detection results at different levels, the comprehensive detection result of seawater under multiple parameters is determined. This comprehensive detection result not only includes the value of a single physical quantity, but also draws a dynamic distribution map of seawater through multi-dimensional cross-validation. At the same time, the spectral shift is further controlled, which improves the accuracy of multiple basic parameters and the accuracy of the comprehensive detection result.

[0082] At this point, multi-source information fusion technology is used to comprehensively process the discrete basic parameters (temperature, salinity, density, pressure, and flow rate) decoupled from step S141 in the data layer and feature layer. By using weighted averaging or Kalman filtering, the optimal estimation of multiple measurements of the same physical quantity is performed, thereby reducing random noise.

[0083] Based on the inherent logical relationship between physical quantities, the fusion process is divided into three sub-detection layers: the basic physical layer, which directly outputs the precise values ​​of temperature, pressure, and flow velocity; the derived parameter layer, which combines temperature and salinity data and uses the seawater state equation (such as the TEOS-10 standard) to calculate seawater density and sound velocity; and the coupling feature layer, which analyzes the correlation between flow velocity and pressure fluctuations and identifies ocean current direction or vortex characteristics. This layered fusion strategy ensures the systematic nature of data processing and avoids the increase in information entropy caused by direct mixing calculations.

[0084] The system utilizes a multi-dimensional cross-validation mechanism to verify the consistency and physical rationality among various parameters. It uses pressure (depth) data to verify the location of the temperature jump layer, or uses the calculated density gradient to verify whether the flow velocity distribution conforms to the geostrophic equilibrium relationship. It employs fuzzy logic or evidence theory to synthesize the evidence from each sub-detection layer. If the physical correlation between a parameter (such as flow velocity) and other parameters (such as pressure gradient) is contradictory, the system will trigger an anomaly alarm and mark the data segment as having reduced credibility. The final output of the comprehensive detection result is a dataset containing numerical values, confidence levels, and status labels.

[0085] The numerical comprehensive detection results are transformed into a visualized spatiotemporal distribution map, which intuitively presents the characteristics of the marine environment. In the time dimension, trend curves of each parameter changing over time are plotted, and periodic features such as tidal cycles and diurnal temperature ranges are marked. In the spatial dimension, temperature-depth profile maps or salinity plane distribution cloud maps are generated by interpolation based on the trajectory data of sensor arrays or mobile platforms. Color mapping is used to render the gradient changes of different parameters, and the current velocity vector field is superimposed to form a dynamic hydrological environment "snapshot". The map includes automatic annotation of key feature nodes (such as thermocline depth and maximum salinity area).

[0086] Specifically, the system performs data fusion on the temperature (+0.5℃) and salinity (+0.9‰) output by S141, correcting minor sensor drift. The first layer confirms the basic environment as "high temperature, high salinity, and low pressure". The second layer, based on the salinity data calibrated according to the "wall thickness-refractive index sensitivity" mapping relationship, combined with the temperature value, calculates that the seawater density has decreased to 1.022 g / cm3, indicating the presence of freshwater inflow. The third layer, combined with the flow velocity data, identifies the presence of vertical density flow in the area. This layered result provides structured data support for subsequent comprehensive judgment.

[0087] For seawater, the system underwent rigorous cross-validation: the system detected an increase in temperature (+0.5℃), which theoretically should be accompanied by an increase in sound speed; simultaneously, it detected an increase in salinity, which would also lead to an increase in sound speed; the system compared the salinity detection data with the calculated sound speed values ​​under the "wall thickness-refractive index sensitivity" mapping relationship and found that the two were highly consistent, verifying the accuracy of detecting small fluctuations in salinity. The final comprehensive detection results were: temperature 25.5℃, salinity 35.9‰, density 1.022g / cm³, with a confidence level of 98%. This result is no longer a list of single values, but an organic whole verified by physical logic, eliminating false salinity signals caused by scattering from suspended particles.

[0088] The system generated a dynamic distribution map of seawater: The map clearly shows the process of salinity slowly increasing from 35.0‰ to 35.9‰ on the time axis, with the curve slope reflecting the rate of evaporation or freshwater inflow. The map uses warm and cool colors to render the temperature field and uses contour lines to mark the density distribution. In particular, for high-sensitivity areas under the "wall thickness-refractive index sensitivity" mapping relationship, the map magnifies and displays the fine-structure changes in density caused by minute salinity fluctuations, exhibiting a "filamentous" distribution characteristic. The map visually shows the presence of a high-temperature, high-salinity water mass invading the seawater monitoring area, accompanied by specific changes in the flow velocity vector field, providing an intuitive and scientific basis for decision-making regarding marine ecological protection and resource development.

[0089] Please see Figure 6 , Figure 6 This is a schematic diagram of the structural composition of a seawater multi-parameter detection system based on fiber optic whispering wall mode according to an embodiment of the present invention; the seawater multi-parameter detection system based on fiber optic whispering wall mode is applied to the above-mentioned seawater multi-parameter detection method based on fiber optic whispering wall mode; the seawater multi-parameter detection system based on fiber optic whispering wall mode includes: The simulation module 21 is used to collect the theoretical analysis model of the whispering wall mode of the fiber optic microbubble cavity during the multi-parameter detection process of seawater, and to perform numerical simulation by combining multi-physics finite element analysis, output multiple simulation results, and determine the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor based on the iteration of multiple simulation results. The parameter change module 22 is used to mark the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor, and to establish a multi-physics coupling model of the microbubble cavity. In the multi-physics coupling model of the microbubble cavity, the temperature gradient field and stress distribution of seawater during transmission in the microbubble cavity are simulated, and the parameter change results of the seawater are marked. The spectral change module 23 is used to determine the spectral change content of seawater based on the fiber optic whispering gallery mode according to the change results of various parameters; analyze the spectral change content and mark the corresponding spectral shift; The integrated detection module 24 is used to determine multiple basic parameters by tracing along the spectral shift, and to determine the integrated detection results of seawater under multiple parameters based on the multiple basic parameters.

[0090] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode, characterized in that, include: In the process of multi-parameter detection of seawater, a theoretical analysis model of the whispering gallery mode of fiber optic microbubble cavity is collected, and numerical simulation is carried out in combination with multiphysics finite element analysis. Multiple simulation results are output, and the influence relationship between the cavity structure parameters of microbubble cavity and the quality factor is determined based on the iteration of multiple simulation results. The influence of the cavity structure parameters of the microbubble cavity on the quality factor was marked, and a multi-physics coupling model of the microbubble cavity was established. In the multi-physics coupling model of the microbubble cavity, the temperature gradient field and stress distribution during seawater transport in the microbubble cavity were simulated, and the changes of various parameters of the seawater were marked. The spectral variations of seawater based on the fiber optic whispering gallery mode were determined based on the results of changes in various parameters. Analyze the spectral changes and mark the corresponding spectral shifts; By tracing the spectral shift, several fundamental parameters are determined, and based on these parameters, a comprehensive detection result of seawater under these parameters is determined.

2. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 1, characterized in that, In the process of multi-parameter detection of seawater, a theoretical analysis model of the whispering gallery mode of the fiber optic microbubble cavity is acquired, and numerical simulation is performed by combining multiphysics finite element analysis. Multiple simulation results are output, and the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor is determined based on the iteration of multiple simulation results, including: The process of real-time monitoring of multi-parameter detection of seawater was carried out, and theoretical analysis models of corresponding fiber optic microcavity whispering wall modes were selected from the seawater detection database. In the theoretical analysis model, the mapping relationship between optical resonance conditions and cavity structure parameters was clarified. At this time, by analyzing the propagation characteristics of light at the interface between optically dense and optically sparse media, the critical incident angle, cavity wall refractive index difference and cavity curvature radius that satisfy the formation of fiber optic microcavity whispering wall modes were determined, and the influence mechanism of geometric morphology on photon trapping ability was clarified.

3. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 2, characterized in that, The process of multi-parameter detection in seawater involves acquiring a theoretical analysis model of the whispering gallery mode of a fiber optic microbubble cavity, combining it with multiphysics finite element analysis for numerical simulation, outputting multiple simulation results, and determining the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor based on the iteration of multiple simulation results. This also includes: Input the cavity structure parameters of the microbubble cavity of seawater, and set the dispersion characteristics, thermo-optic coefficient, and elastic modulus physical properties of fused silica. At the same time, set the corresponding matching layer boundary conditions according to the actual working conditions to construct a highly simulated microcavity geometric model. Input the microcavity geometric model into the optical-fluid coupling module. The optical-fluid coupling module analyzes the disturbance of the evanescent field in the cavity by the microfluidic liquid, and then performs multi-physics finite element analysis on the change of optical resonance characteristics of the microbubble cavity in the fluid environment to output multiple simulation results. The multiple simulation results reflect the interaction between the light field and the external environment. Multiple simulation results are collected and multi-dimensionally matched to output different parameter combinations. Each parameter combination is combined with the quality factor for multi-level iteration, and the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor is determined during the iteration process.

4. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 1, characterized in that, The influence of the cavity structure parameters of the labeled microbubble cavity on the quality factor is established, and a multiphysics coupling model of the microbubble cavity is established. In the multiphysics coupling model of the microbubble cavity, the temperature gradient field and stress distribution during seawater transport in the microbubble cavity are simulated, and the changes in various parameters of the seawater are labeled, including: The influence of the cavity structure parameters of the microbubble cavity on the quality factor was collected, and the influence relationship was analyzed. During the analysis, multiple influence nodes were identified, and the control mechanism of the corresponding parameters was marked for each influence node. Finite element analysis was performed on the multiple influence nodes in different dimensions, and a multiphysics coupling model of the microbubble cavity was gradually constructed during the analysis.

5. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 4, characterized in that, The influence of the cavity structure parameters of the labeled microbubble cavity on the quality factor is established, and a multiphysics coupling model of the microbubble cavity is established. In the multiphysics coupling model of the microbubble cavity, the temperature gradient field and stress distribution during seawater transport in the microbubble cavity are simulated, and the changes of various parameters of the seawater are labeled. The model also includes: A multiphysics coupling model for real-time monitoring of microbubble cavities is developed. This model encompasses the resonant characteristics of the corresponding fiber whispering gallery mode. By introducing optical-mechanical-fluid coupling effects, the physical processes of seawater flowing and filling the microbubble cavity as a microfluidic liquid are further simulated. Combining heat transfer and stress analysis modules, the model simulates the temperature gradient field distribution and stress distribution inside and outside the cavity wall generated during seawater transport within the microbubble cavity. The influence of the effective refractive index and cavity geometry of the microcavity is quantitatively determined, and the changes in various parameters of the seawater, including temperature, pressure, and flow velocity, are marked.

6. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 1, characterized in that, The spectral variation of seawater based on the fiber optic whispering gallery mode is determined according to the results of the changes in various parameters. The spectral changes were analyzed, and the corresponding spectral shifts were marked, including: The results of various parameter changes were collected, and the intrinsic mapping mechanism between seawater parameter changes and spectral drift was determined by combining the characteristics of spectral drift. In this mapping mechanism, the generated transmission spectrum data were analyzed, and the dynamic evolution of the center wavelength position, peak intensity and spectral linewidth of the echo-gallery mode resonance peak was monitored. The spectral feature change patterns caused by different seawater parameters were identified, the corresponding spectral change content was captured, and the characteristics of spectral morphology change with the seawater environment were presented.

7. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 6, characterized in that, The spectral variation of seawater based on the fiber optic whispering gallery mode is determined according to the results of the changes in various parameters. The analysis of this spectral change, marking the corresponding spectral shift, also includes: In this spectral change content, the spectrum after environmental disturbance is marked, the reference spectrum and the spectrum after environmental disturbance are compared, the drift of the resonant wavelength is extracted, the drift of the resonant wavelength is used as the core sensing signal, and the spectral response differences of different order modes are distinguished to be converted into quantifiable spectral shift, and a quantitative correspondence between spectral shift and changes in multiple parameters of seawater is established.

8. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 1, characterized in that, The process of tracing along the spectral shift to determine multiple fundamental parameters, and then determining the comprehensive detection results of seawater under these parameters, includes: The spectral shift was traced, and the corresponding spectral shift differences were determined during the tracing process. The basic parameters were determined by matrix inversion, and the independent contribution of each basic parameter to the spectral shift was marked. Deep learning was performed on each basic parameter, and cross-sensitive interference factors in the measurement of a single parameter were eliminated during the learning process. The basic parameters cover the temperature, salinity, density, pressure and flow velocity of seawater.

9. The method for detecting multiple parameters of seawater based on fiber optic whispering gallery mode according to claim 8, characterized in that, The process of determining multiple basic parameters by tracing along the spectral shift, and determining the comprehensive detection result of seawater under multiple parameters based on these basic parameters, further includes: The data of each basic parameter is fused, and the sub-detection results at different levels are divided in sequence during the fusion process. Based on the multiple sub-detection results at different levels, the comprehensive detection result of seawater under multiple parameters is determined. This comprehensive detection result not only includes the value of a single physical quantity, but also draws a dynamic distribution map of seawater through multi-dimensional cross-validation.

10. A multi-parameter seawater detection system based on fiber optic whispering gallery mode, characterized in that, The seawater multi-parameter detection system based on fiber optic whispering wall mode is applied to the seawater multi-parameter detection method based on fiber optic whispering wall mode as described in any one of claims 1-9. The seawater multi-parameter detection system based on fiber optic whispering gallery mode includes: The simulation module is used to acquire the theoretical analysis model of the whispering wall mode of fiber optic microbubble cavity during the multi-parameter detection process of seawater, and to perform numerical simulation by combining multiphysics finite element analysis, outputting multiple simulation results, and determining the influence relationship between the cavity structure parameters of the microbubble cavity and the quality factor based on the iteration of multiple simulation results. The parameter variation module is used to mark the influence of the cavity structure parameters of the microbubble cavity on the quality factor, and to establish a multi-physics coupling model of the microbubble cavity. In the multi-physics coupling model of the microbubble cavity, the temperature gradient field and stress distribution of seawater during transport in the microbubble cavity are simulated, and the results of various parameter variations of seawater are marked. The spectral variation module is used to determine the spectral variation of seawater based on the fiber optic whispering gallery mode according to the variation results of various parameters; analyze the spectral variation and mark the corresponding spectral shift. The integrated detection module is used to determine multiple basic parameters by tracing along the spectral shift, and to determine the integrated detection results of seawater under multiple parameters based on these basic parameters.

Images