Airport pavement overlay thickness calibration method, system and storage medium

By combining multi-frequency electrical parameters with thermodynamic parameters, a three-dimensional coupled correction model is constructed to dynamically correct the thickness of the airport pavement medium covering layer. This solves the problems of large thickness measurement errors and false alarms in low-temperature environments, and enables accurate measurement of water, ice, and snow thickness, supporting all-weather safe operation and maintenance of airports.

CN123065331APending Publication Date: 2026-11-10CIVIL AVIATION UNIV OF CHINA +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611474497.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2026-09-09
Filing Date
2026-09-21
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing airport pavement medium cover thickness detection equipment suffers from large errors, false alarms, and missed alarms in low-temperature environments. It cannot accurately measure the thickness of water accumulation, de-icing liquid film, ice accumulation, and snow accumulation, especially under the combined effects of temperature and salinity, the dielectric constant deviates significantly due to nonlinear shift.

Method used

By combining multi-frequency electrical parameters with environmental thermodynamic parameters, the phase state of the dielectric coating layer is identified, and a three-dimensional coupled correction model of temperature-salinity-dielectric thickness is constructed to dynamically correct the dielectric coating layer thickness, eliminate temperature drift and salinity interference, and optimize impedance weights through a multilayer perceptron classification model and Debye relaxation characteristics to achieve accurate thickness measurement.

Benefits of technology

Accurately measure the thickness of the medium covering layer under complex operating conditions, reduce misjudgments and omissions, and provide reliable data support for airport de-icing and snow maintenance and aircraft take-off and landing safety assessments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123065331A_ABST
    Figure CN123065331A_ABST
Patent Text Reader

Abstract

This invention relates to a method, system, and medium for calibrating the thickness of airport pavement media cover. The method includes: acquiring multi-frequency electrical parameters and environmental thermodynamic parameters using a media sensor; obtaining the real-time salt concentration of the medium based on the multi-frequency electrical parameters; identifying the phase type of the current pavement media cover by combining the multi-frequency electrical parameters and the real-time pavement temperature; configuring corresponding multi-frequency point combinations and impedance weights for different medium phase differences; constructing a three-dimensional coupled correction model of temperature-salinity-media thickness to obtain multi-parameter coupling errors; and dynamically correcting the original thickness of the media cover based on the impedance weights corresponding to the medium phase and the multi-parameter coupling errors, eliminating thickness measurement deviations caused by temperature drift, salinity interference, and phase differences, and outputting the true physical thickness of the airport pavement media cover. This invention can accurately obtain the true physical thickness of accumulated water film, de-icing film, ice accumulation, and snow accumulation, reducing monitoring misjudgments and omissions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of aviation ground support and intelligent transportation sensing technology, specifically relating to a method, system and storage medium for calibrating the thickness of airport pavement medium covering layer. Background Technology

[0002] In low-temperature winter conditions, airport pavements are prone to forming media covering layers such as water film, de-icing liquid film, ice, and snow. The thickness of these media covering layers directly determines the runway friction coefficient and is a core basis for airport de-icing and snow maintenance operations and ensuring the safety of aircraft take-off and landing. Therefore, it is necessary to accurately measure the true thickness of various media covering layers on the pavement.

[0003] Existing dielectric detection equipment relies on sensors to collect electrical signals at a single frequency or fixed multiple frequencies, and directly calculates the thickness of the medium using a fixed conversion model. On the one hand, the dielectric polarization response characteristics of different media such as water, de-icing fluid, ice, and snow vary significantly, but existing equipment uses the same set of frequencies and impedance weights to calculate the thickness of all media. On the other hand, existing technologies only use a simple single-variable linear compensation method, resulting in a large systematic error in the calculated medium thickness.

[0004] In addition, existing equipment is unable to quantify the nonlinear shift in dielectric constant caused by the combined effects of temperature and salinity, resulting in severe distortion of the final output thickness value and frequent issues of missed alarms and false alarms, which cannot meet the airport's need for accurate all-weather monitoring of pavement ice and snow conditions. Summary of the Invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, the present invention proposes a method, system and storage medium for calibrating the thickness of the airport pavement medium cover layer.

[0006] This invention provides a method for calibrating the thickness of the airport pavement medium cover layer, comprising:

[0007] Multi-frequency electrical parameters and environmental thermodynamic parameters are collected using media sensors deployed on the airport pavement. The multi-frequency electrical parameters include preset multi-frequency impedance signals, and the environmental thermodynamic parameters include real-time pavement temperature. Real-time salt concentration of the media is obtained based on the multi-frequency electrical parameters. The phase type of the current pavement media cover layer is identified by combining the multi-frequency electrical parameters and the real-time pavement temperature. The media cover layer includes water film, de-icing film, ice accumulation, and snow accumulation. Corresponding multi-frequency combinations and impedance weights are configured for different media phases. The original uncorrected thickness of the media cover layer is initially calculated based on the preset multi-frequency impedance signals. Based on multi-frequency electrical parameters, real-time pavement temperature, real-time medium salt concentration, original thickness of the medium cover layer, medium phase identification, and Debye relaxation intrinsic parameters corresponding to various media, a three-dimensional coupled correction model of temperature-salinity-medium thickness is constructed. The nonlinear offset of the complex permittivity of the medium under different operating conditions is calculated to obtain the multi-parameter coupling error. Based on the impedance weight corresponding to the medium phase and the multi-parameter coupling error, the original thickness of the medium cover layer is dynamically corrected to eliminate the thickness measurement deviation caused by temperature drift, salinity interference, and phase difference, and output the true physical thickness of the airport pavement medium cover layer.

[0008] In one optional embodiment, the water film and the de-icing film use frequencies of 20kHz and 100kHz, respectively, and the ice and the snow use frequencies of 5kHz, 10kHz, and 100kHz, respectively.

[0009] In one alternative implementation, a thickness conversion relationship is established based on the Debye relaxation mechanism and the sensor edge electric field distribution constant. Corresponding frequency points are matched for different medium phase states and impedance weighting factors are adaptively allocated. Independent nonlinear temperature compensation factors are configured for various phase states. After correcting the nonlinear distortion of the electrical signal generated by the polarization of the electrode, the medium thickness is calculated.

[0010] In one optional implementation, the real-time salt concentration of the medium is obtained by inverting the multi-frequency electrical parameters as follows: the amplitude and phase of the multi-frequency impedance are substituted into the pre-calibrated salinity response mapping curve, and the real-time salt concentration of the medium is obtained by solving.

[0011] In one optional implementation, a multilayer perceptron classification model is used to identify the phase state type of the medium covering layer, with the multi-frequency electrical parameters, the real-time pavement temperature, and the real-time salt concentration of the medium as model input features.

[0012] In one optional implementation, the Debye relaxation intrinsic parameters corresponding to the various media include the static dielectric constant, the high-frequency limiting dielectric constant, and the dipole relaxation time.

[0013] In one optional implementation, the temperature-salinity-medium thickness three-dimensional coupled correction model uses the complex permittivity offset as an intermediate variable to establish a nonlinear correlation surface among temperature, salt concentration, and medium thickness.

[0014] In one optional implementation, after dynamically correcting the original thickness of the dielectric cover layer, the corrected actual physical thickness, corresponding dielectric phase, and real-time operating parameters are synchronously stored in a local time series database.

[0015] On the other hand, a system for calibrating the thickness of an airport pavement medium cover layer includes: a parameter acquisition module for acquiring multi-frequency electrical parameters and environmental thermodynamic parameters of the airport pavement, wherein the multi-frequency electrical parameters include preset multi-frequency impedance signals, and the environmental thermodynamic parameters include real-time pavement temperature; a salinity inversion and phase identification module for inverting the real-time salt concentration of the medium based on the multi-frequency electrical parameters, and identifying the phase type of the current pavement medium cover layer by combining the multi-frequency electrical parameters and the real-time pavement temperature, wherein the medium cover layer includes water film, de-icing film, ice accumulation, and snow accumulation; and a frequency weight configuration and original thickness calculation module for calculating the thickness of different medium phases. The system uses a multi-frequency combination and impedance weighting corresponding to the heterogeneous configuration, and preliminarily calculates the uncorrected original thickness of the dielectric overlay based on the preset multi-frequency impedance signal. A three-dimensional coupling adaptive calibration module is used to combine the multi-frequency electrical parameters, the real-time pavement temperature, the real-time salt concentration of the medium, the original thickness of the dielectric overlay, the medium phase identifier, and the Debye relaxation intrinsic parameters corresponding to various media to construct a three-dimensional coupling correction model of temperature-salinity-medium thickness. The module calculates the multi-parameter coupling error and dynamically corrects the original thickness of the dielectric overlay based on the impedance weight corresponding to the medium phase and the multi-parameter coupling error, outputting the true physical thickness of the airport pavement dielectric overlay.

[0016] The present invention also proposes a storage medium storing a computer program, which, when executed by a processor, implements any of the airport pavement medium overlay thickness calibration methods described in the present invention.

[0017] The beneficial technical effects of this invention are as follows: This invention collects multi-frequency electrical parameters and real-time pavement temperature, inverts the real-time salt concentration of the medium based on the multi-frequency electrical parameters, and identifies the phase state of the medium covering layer. It then configures multi-frequency point combinations and impedance weights differently for different medium phase states; constructs a three-dimensional coupled correction model of temperature-salinity-medium thickness, combines the nonlinear shift of the complex permittivity with the Debye relaxation intrinsic parameters of various media, obtains the multi-parameter coupling error, and dynamically corrects the original thickness of the medium based on the impedance weights and coupling error. This invention fully considers the multiple coupling interferences of temperature, de-icing fluid salinity, and medium phase state, matches the dielectric polarization response characteristics of different covering media, eliminates thickness measurement deviations caused by the combined effects of multiple factors, and can accurately obtain the true physical thickness of accumulated water film, de-icing fluid film, ice accumulation, and snow accumulation under complex working conditions such as rain, snow, de-icing, and de-icing fluid spraying on airport pavements, reducing monitoring misjudgments and omissions, and providing accurate data for airport runway de-icing and snow maintenance and takeoff and landing safety assessment. Attached Figure Description

[0018] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0019] Figure 1 A flowchart illustrating an embodiment of the airport pavement medium overlay thickness calibration method provided by the present invention;

[0020] Figure 2 A schematic diagram illustrating the principle of an airport pavement medium overlay thickness calibration method provided in an embodiment of the present invention;

[0021] Figure 3 A schematic diagram of the hardware structure of a multifunctional passive pavement sensor probe for an airport pavement medium cover layer thickness calibration method provided in an embodiment of the present invention;

[0022] Figure 4 A schematic diagram of the machine learning model architecture for pavement condition identification and multi-field coupled parametric regression in the airport pavement medium overlay thickness calibration method provided in an embodiment of the present invention;

[0023] Figure 5 A schematic diagram illustrating the data storage, early warning, and distributed management module flow of the airport pavement medium cover layer thickness calibration method provided in an embodiment of the present invention;

[0024] Figure 6 A comparison diagram of clustering and adaptive calibration of broadband electrical discrete characteristics of the cover material under different salinities in an embodiment of the present invention for the airport pavement medium cover thickness calibration method provided by the present invention.

[0025] Figure 7Salinity response curves of dual-frequency impedance modulus ratio under different cover thicknesses for the airport pavement medium cover thickness calibration method provided in an embodiment of the present invention;

[0026] Figure 8 The dual-frequency impedance characteristic response surface diagram under temperature and salinity coupling for the airport pavement medium overlay thickness calibration method provided in an embodiment of the present invention;

[0027] Explanation of reference numerals in the attached diagram: 1-High-strength corrosion-resistant encapsulation base; 2-Multi-frequency measurement electrode array 2; 3-High-precision surface temperature probe; 4-Computing motherboard; 5-Industrial bus interface. Detailed Implementation

[0028] The purpose of this invention is to solve the key technical problems of traditional passive pavement sensors, such as insufficient accuracy, difficulty in identifying the type of cover, and distortion in thickness measurement caused by physical reference offset under complex temperature changes and de-icing fluid environments.

[0029] The following is in conjunction with the appendix Figures 1 to 8 The present application will be further described in detail with reference to the embodiments. It is understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0030] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0031] Please refer to Figure 1 and Figure 2 This invention provides a method for calibrating the thickness of the airport pavement medium cover layer, comprising the following steps:

[0032] Step S101: Use medium sensors deployed on the airport pavement to collect multi-frequency electrical parameters and environmental thermodynamic parameters. The multi-frequency electrical parameters include preset multi-frequency impedance signals, and the environmental thermodynamic parameters include the real-time pavement temperature.

[0033] Step S102: Based on the inversion of multi-frequency electrical parameters, the real-time salt concentration of the medium is obtained. Combined with the multi-frequency electrical parameters and the real-time temperature of the pavement, the phase type of the current pavement medium covering layer is identified. The medium covering layer includes water film, de-icing film, ice accumulation, and snow accumulation.

[0034] Step S103: Configure corresponding multi-frequency combinations and impedance weights for different medium phase states;

[0035] Step S104: Based on the preset multi-frequency impedance signal, the original thickness of the uncorrected dielectric cover layer is initially calculated;

[0036] Step S105: Based on multi-frequency electrical parameters, real-time pavement temperature, real-time medium salt concentration, original thickness of medium cover layer, medium phase identification, and Debye relaxation intrinsic parameters corresponding to various media, construct a three-dimensional coupled correction model of temperature-salinity-medium thickness, calculate the nonlinear offset of the complex permittivity of the medium under different working conditions, and obtain the multi-parameter coupling error.

[0037] Step S106: Based on the impedance weight corresponding to the phase state of the medium and the multi-parameter coupling error, the original thickness of the medium cover layer is dynamically corrected to eliminate the thickness measurement deviation caused by temperature drift, salinity interference and phase state differences, and output the true physical thickness of the airport pavement medium cover layer.

[0038] like Figure 3 As shown, the all-weather pavement ice and snow sensing device of this invention adopts a flush-mount design for installation on the pavement, and includes the following main components: A high-strength, corrosion-resistant encapsulation base 1 is made of special engineering plastic, possessing excellent pressure resistance and corrosion resistance, adapting to long-term vehicle traffic on the runway. A multi-frequency measurement electrode array 2 is embedded in the top surface of the sensor, containing a ring-shaped arrangement of corrosion-resistant metal alloy electrodes, used to transmit and receive AC broadband excitation signals to form an edge-penetrating electric field. A high-precision surface temperature probe 3 is encapsulated in the same layer as the electrodes, enabling real-time sensing of the actual temperature gradient of the surface layer. An edge computing motherboard 4 is integrated within the bottom sealed chamber, encompassing a precision impedance conversion unit, an analog-to-digital conversion module, and a core processing chip. A waterproof communication and industrial bus interface 5 extends from the bottom of the sensor, supporting RS485 industrial bus and Modbus RTU protocols, ensuring stable and reliable data transmission and system power supply even under harsh winter geological conditions.

[0039] Multi-dimensional collaborative data acquisition is specifically configured as follows: a passive pavement sensor probe is embedded in the pavement. Through the built-in high-precision impedance conversion circuit and microcontroller, discrete frequency point excitation is performed within the scanning frequency range to synchronously acquire the amplitude and phase information of the corresponding frequency point. Environmental parameters are fused by combining the thermodynamic sensor array of pavement surface and depth.

[0040] This invention uses a medium sensor to simultaneously collect multi-frequency electrical parameters and real-time pavement temperature. It relies on multi-frequency impedance amplitude and phase information to fully capture the dielectric polarization characteristics of different media, providing rich basic features for subsequent salt concentration inversion and medium phase identification.

[0041] This method calibrates the type identification boundary by analyzing the sensitivity of temperature and salinity to the type of cover under multi-field coupling environment. Based on the real-time salt concentration of the medium obtained by inverting multi-frequency electrical parameters, and combined with electrical parameters and temperature, it achieves phase identification of water film, de-icing film, ice accumulation, and snow accumulation. It fully considers the changes in dielectric properties caused by the salinity of the de-icing fluid, and can distinguish between salt-containing films and pure water or ice layers, avoiding misidentification of the medium type.

[0042] The system simultaneously acquires temperature input data, multi-frequency impedance parameter input data, and ambient temperature and humidity parameters from pavement sensors, and extracts feature parameters. Through simulation studies and experiments, it verifies the sensitivity of temperature and salinity to different types of pavement coverings, extracts feature parameters with significant impact on the identification results, and achieves benchmark calibration of the pavement covering type identification boundary. The extracted feature parameters are input into a pre-trained machine learning-based salinity-type calibration model. The salinity-type calibration model, combined with dielectric polarization mechanism, identifies and outputs the specific type classification result of the current pavement covering, including dry, wet, waterlogged, ice, and snow.

[0043] This method, based on the known type of covering, can eliminate measurement errors caused by differences in liquid concentration and temperature drift.

[0044] By analyzing the dielectric response characteristics of different physical states (water, ice, snow, frost), multi-dimensional feature vectors such as impedance amplitude ratio, phase difference, and dielectric relaxation characteristics are extracted. A pre-trained salinity-type calibration model is used to achieve automatic and accurate identification of pavement dryness, wetness, water accumulation, ice accumulation, and snow accumulation.

[0045] By analyzing the polarization characteristics of polar molecules such as water, ice, and snow at different AC frequencies, multidimensional feature vectors, including impedance amplitude ratio at characteristic frequencies, phase characteristic difference, and dielectric relaxation parameters, are extracted. Then, an improved multilayer perceptron (MLP) machine learning classification algorithm is used to establish a pavement physical state identification model, achieving accurate classification of pavement coverings under complex working conditions.

[0046] The extraction process involves a multi-dimensional feature vector, including the impedance amplitude ratio at characteristic frequencies, phase characteristic difference, dielectric loss tangent, and broadband electrical discrete parameters used to characterize phase state boundaries. The specific calculation formula is as follows:

[0047] ;

[0048] ;

[0049] ;

[0050] ;

[0051] in, and The impedance amplitudes are at 100kHz and 20kHz, respectively. and This represents the AC impedance phase angle at the corresponding frequency point; Let be the real part of the complex permittivity. This represents the imaginary part of the complex permittivity. Represents the discrete coefficients over a wide frequency range. and They represent low frequencies within a wide frequency range. With high frequency The measured values ​​of electrical parameters are shown below. For the complex transitional phases between brine and ice, this dispersion coefficient can significantly amplify the low-frequency space charge polarization characteristics of the physical medium.

[0052] Based on novel experimental calibration data obtained under different temperature ranges and salt concentrations, the curvature of the dielectric constant at characteristic frequency points was established; it was clarified that the surface temperature gradient and broadband electrical discrete parameters are mainly used to construct phase boundaries for type identification. The system uses broadband AC impedance sweep frequency technology to obtain the continuous electrical response spectrum within the target frequency band, and extracts the broadband capacitance and conductance discrete coefficients accordingly.

[0053] The extracted feature parameters are input into the temperature-salinity-thickness calibration model. This model establishes a multidimensional coupling mapping by combining the dielectric polarization mechanism and Debye relaxation characteristics, outputting the corresponding baseline salinity and temperature correction values. Based on the current pavement overlay classification results, the temperature compensation factor and multi-frequency impedance weights are adaptively adjusted. Furthermore, using the corrected accurate salt concentration, the dielectric constant and conductivity parameters in the thickness measurement benchmark are dynamically updated.

[0054] A multidimensional coupling mapping between characteristic parameters and salt concentration and thickness is established. Specifically, the Debye relaxation characteristic equation is introduced to analyze the polarization response delay of dipoles inside the covering material under an alternating electric field. The complex permittivity model formula is as follows:

[0055] ;

[0056] In the formula, The complex permittivity under multi-field coupling conditions, The static dielectric constant is The dielectric constant of optical frequency, For polarization relaxation time, Let be the angular frequency of the alternating electric field. By fitting the frequency shift characteristics of the dielectric loss peak under different salt concentrations, the real and imaginary parts of the complex permittivity are deconstructed, thereby outputting high-precision baseline salinity and temperature correction values.

[0057] An influence weight analysis was performed on the extracted parameters, and the impedance modulus ratio and phase difference, which are extremely sensitive to the ice and snow phase state, were extracted as the core parameters. Then, a weighted feature fusion strategy was adopted to construct a comprehensive feature vector formula:

[0058] ;

[0059] in: The resulting multidimensional feature vector; These are the corresponding adaptive fusion weights. These represent surface temperature gradient characteristics, electrical discrete characteristics, polarization relaxation characteristics, and impedance comprehensive characteristics, respectively. These fused characteristics, along with the described change in curvature, serve as the basis for determining polarization features in the machine learning physical classification model, thereby correcting and converging the boundary recognition thresholds for different physical phases.

[0060] The extracted multidimensional feature vectors and their corresponding pavement states are used as the training set for machine learning, and a weighted feature fusion strategy is employed. Specifically, each principal feature parameter is assigned a different weight based on its contribution to ice and snow type identification, and these weights are fused to form a weighted feature vector. The weighted feature fusion strategy is defined as follows:

[0061] ;

[0062] in, This represents the combined characteristics of impedance amplitude and phase. This represents the comprehensive characteristics of dielectric relaxation. The normalized temperature gradient characteristic; These are the adaptively assigned fusion weight coefficients.

[0063] Based on the identified physical state, a temperature-salinity-thickness calibration model is used to calculate the precise salt concentration and actual freezing point of the current liquid film or ice / snow. Simultaneously, a nonlinear temperature compensation factor and a differentiated multi-feature frequency impedance weighting mechanism are introduced for different cover conditions to dynamically compensate the thickness measurement benchmark, ultimately accurately calculating the physical thickness.

[0064] A nonlinear conversion relationship between the electrical response signal and the actual dielectric thickness is established. Based on the dielectric response principle, the impedance weighting factor at the characteristic frequency point is adaptively extracted and updated. Combined with dynamic weight allocation, the physical thickness of ice and water film is accurately calculated.

[0065] Based on the different phase states of different media, multi-frequency point combinations and impedance weights are configured to match the polarization response laws of different media such as water, ice and snow under alternating electric fields, so that the thickness calculation can fully adapt to the electrical characteristics of the medium itself.

[0066] The technical solution of this invention is well-suited to the all-weather operation and maintenance scenario of airport pavements. It can stably and accurately measure the thickness of the medium covering layer under complex working conditions such as rain, snow, ice melting, and spraying de-icing fluid, effectively reducing the probability of misjudgment and omission in pavement condition monitoring, and providing reliable data support for airport de-icing and snow removal scheduling and runway operation safety assessment.

[0067] Furthermore, the water accumulation film and the de-icing film are selected at frequencies of 20kHz and 100kHz, respectively, while the ice accumulation and snow accumulation are selected at frequencies of 5kHz, 10kHz, and 100kHz, respectively.

[0068] The extracted feature parameters specifically include: surface temperature gradient, impedance weighting factors at characteristic frequency points of 5kHz, 10kHz, 20kHz and 100kHz, and capacitance and conductance dispersion coefficients over a wide frequency range.

[0069] The adaptive extraction and updating of impedance weighting factors specifically involves: defining impedance weighting factors at specific frequencies primarily for thickness calculation; performing algorithm iterations based on calibration data; and constructing differentiated characteristic frequency response mappings for different physical states of the covering material to optimize the impedance weighting factors: when the covering material is in a waterlogged state, the impedance weighting factors at 20kHz and 100kHz characteristic frequencies are extracted and optimized; when the covering material is in an icy state, the impedance weighting factors at 5kHz, 10kHz, and 100kHz characteristic frequencies are extracted and optimized; simultaneously, for different states such as icing or waterlogging, independent temperature compensation factors are dynamically introduced to correct the conductivity parameter in the edge electric field model, eliminate nonlinear offsets, and achieve accurate thickness calculation.

[0070] The core of the dynamic correction and thickness calculation process lies in eliminating measurement benchmark distortion caused by the decrease in salt concentration during winter de-icing and drastic temperature fluctuations. A multi-dimensional coupled mapping between feature parameters and salt concentration and thickness is established through a machine learning regression model, outputting baseline salinity and temperature correction values.

[0071] Specifically, the actual thickness is calculated by combining the edge electric field calibration parameters. Based on the principle of electromagnetic edge field penetration depth, the measured capacitance change is converted into physical thickness. Since the dielectric constant is highly susceptible to the effects of temperature fluctuations in water and ice accumulation, this invention replaces the static physical benchmark with a temperature-salinity multidimensional coupling model and constructs differentiated characteristic frequency response mappings for different physical states of the covering material to optimize the impedance weighting factor: when identified as a water accumulation state, the impedance weights at 20kHz and 100kHz characteristic frequencies are extracted and optimized; when identified as an ice accumulation state, the impedance weights at 5kHz, 10kHz, and 100kHz characteristic frequencies are extracted and optimized. The final high-precision thickness calculation formula is:

[0072] ;

[0073] in, The corrected physical thickness. The physical size transformation constant of the electric field at the sensor edge is... For capacitance measurement, For passing temperature With precise salt concentration Real-time dynamic dielectric constant after dynamic compensation. The adaptive impedance weighting coefficients are assigned differently for specific states as described above. The impedance measured at the corresponding frequency point. This is a nonlinear temperature compensation factor for frost and ice.

[0074] When the covering is in a state of water accumulation, characteristic frequency points Taking 20kHz and 100kHz as examples, when the covering is in an icy state, the characteristic frequency points are... We used 5kHz, 10kHz, and 100kHz frequencies to correct the conductivity parameter in the edge electric field model, eliminate nonlinear offsets, and achieve accurate thickness calculation.

[0075] Figure 7 The curves are two-dimensional characteristic response curves. The horizontal axis represents the conductivity of the water medium (sigma, characterizing salinity), and the vertical axis represents the impedance modulus ratio of the 20kHz / 100kHz frequency bands |Z|_{20kHz} / |Z|_{100kHz}. The four curves correspond to cover thicknesses of 0.5mm, 1.0mm, 2.0mm, and 4.0mm, respectively.

[0076] The overall trend is as follows: with a fixed cover thickness, the impedance modulus ratio decreases monotonically with increasing medium salinity (conductivity); under the same salinity conditions, the greater the cover thickness, the lower the overall value of the dual-frequency impedance modulus ratio. This figure visually demonstrates that salinity and thickness simultaneously alter the multi-frequency impedance characteristics, and traditional fixed-weight thickness measurement methods cannot distinguish between the interference of the two. It also corroborates the necessity of the differentiated frequency weighting and salinity pre-correction in this invention, providing experimental data support for adaptive impedance weight allocation and thickness calculation formulas.

[0077] Figure 8 The surface is a three-dimensional coupled response surface. The X-axis represents the salinity of the water, the Y-axis represents the ambient temperature, and the Z-axis represents the impedance modulus ratio of the 20kHz / 100kHz dual-frequency circuit. The color gradient of the surface represents the magnitude of the impedance modulus, with the yellow area representing a high impedance ratio and the dark blue area representing a low impedance ratio.

[0078] The surface change trend shows that both increased salinity and increased temperature will synergistically reduce the dual-frequency impedance modulus. There is strong nonlinear coupling interference between the two, and single temperature compensation or single salinity compensation cannot completely offset the measurement offset. The temperature-salinity-thickness three-dimensional coupling calibration model constructed in this invention can completely fit the nonlinear change law of the surface and simultaneously correct the dielectric constant drift caused by the two types of variables, thus solving the problem of temperature-salinity coupling error of traditional sensors from a mechanistic perspective.

[0079] Based on the Debye relaxation mechanism and the sensor edge electric field distribution constant, a thickness conversion relationship is established. Corresponding frequency points are matched for different medium phase states and impedance weighting factors are adaptively allocated. Independent nonlinear temperature compensation factors are configured for various phase states. After correcting the nonlinear distortion of electrical signals caused by plate polarization, the medium thickness is calculated.

[0080] By acquiring reference data under a standard cover with known thickness and condition, the initial response parameters and edge electric field distribution constant of the sensor probe are determined; and a temperature compensation factor is introduced in the calculation process to dynamically correct the inherent nonlinear electrical signal distortion caused by drastic fluctuations in ambient temperature and polarization of the measuring electrode surface.

[0081] The method for obtaining the real-time salt concentration of the medium based on the inversion of multi-frequency electrical parameters is as follows: substitute the multi-frequency impedance amplitude and phase into the pre-calibrated salinity response mapping curve, and solve to obtain the real-time salt concentration of the medium.

[0082] like Figure 4 As shown, a multilayer perceptron classification model is used to identify the phase state type of the medium covering layer, with multi-frequency electrical parameters, real-time pavement temperature, and real-time medium salt concentration as the model input features.

[0083] Figure 4 This paper demonstrates the overall architecture of an improved MLP (Multilayer Perceptron) model for pavement cover type identification and salinity regression calculation. The model input receives a weighted and fused multidimensional feature vector, including surface temperature gradient, multi-frequency impedance amplitude / phase, wideband capacitance / conductivity discrete coefficients, and Debye relaxation dielectric features. The intermediate hidden layer performs nonlinear feature mapping and adaptive iterative training of the weights. The output layer has two branches: one for classification, outputting phase labels for five types of pavement cover—dry, wet, waterlogged, icy, and snowy—corresponding to the salinity-type calibration model; the other for regression, outputting real-time salt concentration correction parameters for use in the temperature-salinity-thickness calibration model. This architecture is built upon the physical mechanism of dielectric polarization, unlike purely black-box machine learning models without physical constraints, and can achieve accurate phase classification and salinity calculation under complex temperature-salinity coupling conditions.

[0084] Figure 6This is a two-dimensional scatter plot, with the horizontal axis representing the capacitance dispersion coefficient D_C and the vertical axis representing the conductivity dispersion coefficient D_G. The legend distinguishes six types of pavement covering sample points: dry, pure water, snow, ice, and low / high salinity de-icing fluid. The black solid line in the figure is the macroscopic baseline boundary between pure water and snow / ice, the red dashed line is the salinity drift adaptive calibration fitting line of this invention, and the blue dashed line is the phase fine segmentation boundary output by the machine learning model.

[0085] The comparison shows that before calibration, there is a large area of ​​overlapping clusters between the salt-containing de-icing fluid samples and the pure water and ice samples, which can easily lead to misjudgment of phase state. After the present invention introduces temperature and salinity sensitivity weight correction, the cluster intervals of various media are converged through adaptive calibration curves. The boundaries of high-salt, low-salt, pure water, ice and snow sample clusters are clearly separated, which verifies that the boundary calibration and weighted feature fusion scheme can effectively eliminate salinity interference and improve the classification accuracy of cover.

[0086] The inherent parameters of Debye relaxation for various media include the static dielectric constant, the high-frequency limiting dielectric constant, and the dipole relaxation time.

[0087] A three-dimensional coupled correction model of temperature-salinity-medium thickness is established using the complex permittivity offset as an intermediate variable to create a nonlinear correlation surface among temperature, salt concentration, and medium thickness.

[0088] Furthermore, such as Figure 5 As shown, after dynamically correcting the original thickness of the medium cover layer, the corrected actual physical thickness, corresponding medium phase state, and real-time operating parameters are synchronously stored in the local time series database.

[0089] As a back-end processing step in the detection process, the time-series database receives the precise cover type and thickness values ​​after calculation, establishes a real-time monitoring database based on time series, and synchronously records the corresponding detection time information.

[0090] Figure 5 The invention fully presents the entire data flow of backend data processing, security assessment, and alarm push: the multi-frequency impedance and thermodynamic parameters collected by the front-end multi-dimensional sensing module are solved by the model and output the time series data of the cover type, thickness, and salinity; the time series database completes the persistent storage of data with timestamps and equipment location information; the threshold judgment unit retrieves the preset runway friction safety threshold to complete the pavement risk assessment; if the detected data exceeds the safety threshold, the comprehensive monitoring report integrating ice and snow type, thickness, and salinity will be synchronously sent to the two dispatch systems of airport ground maintenance and snow and ice removal support through the wireless network to realize the closed loop of all-weather automatic monitoring and active alarm.

[0091] Specifically, the system imports the real-time stored cover type and thickness values ​​into the built-in threshold determination unit, dynamically compares them with preset road safety thresholds, and performs a road condition safety assessment; it integrates the ice and snow type identification results and thickness measurement data to generate a real-time comprehensive report; when the step-by-step determination data exceeds the safety threshold, it pushes the comprehensive report and alarm information to the ground maintenance management system and support system simultaneously through the wireless network communication interface.

[0092] A sequence database is used to store edge node data. The data table structure includes fields such as timestamp, device ID, detection node location, pavement physical condition, precise thickness, salinity value, and environmental parameters.

[0093] It also includes a data management and scheduling interaction module, used to comprehensively process the above results to establish a distributed early warning platform, and to push alarm commands to the airport management system in real time when the safety threshold is exceeded. In this embodiment, a pavement sensor probe with a passive measurement architecture is used, employing wideband AC impedance sweep frequency technology, a high-precision surface temperature probe, and an environmental temperature and humidity sensing unit.

[0094] Simultaneously acquire low-level electromagnetic and thermodynamic data under complex coupling environments.

[0095] On the other hand, this invention also proposes a system for calibrating the thickness of the airport pavement medium cover layer, comprising: a parameter acquisition module, a salinity inversion and phase identification module, a frequency point weighting and original thickness calculation module, and a three-dimensional coupled adaptive calibration module. The parameter acquisition module is used to acquire multi-frequency electrical parameters and environmental thermodynamic parameters of the airport pavement. The multi-frequency electrical parameters include preset multi-frequency impedance signals, and the environmental thermodynamic parameters include the real-time pavement temperature. The salinity inversion and phase identification module is used to obtain the real-time salt concentration of the medium based on the multi-frequency electrical parameters, and to identify the phase type of the current pavement medium cover layer by combining the multi-frequency electrical parameters and the real-time pavement temperature. The medium cover layer includes water film, de-icing film, ice accumulation, and snow accumulation. The point weight configuration and original thickness calculation module is used to configure corresponding multi-frequency point combinations and impedance weights for different media phase states, and to preliminarily calculate the uncorrected original thickness of the media cover layer based on the preset multi-frequency impedance signals. The three-dimensional coupling adaptive calibration module is used to combine multi-frequency electrical parameters, real-time pavement temperature, real-time medium salt concentration, original thickness of the media cover layer, medium phase state identifier, and Debye relaxation intrinsic parameters corresponding to various media to construct a temperature-salinity-media thickness three-dimensional coupling correction model, calculate the multi-parameter coupling error, and dynamically correct the original thickness of the media cover layer according to the impedance weight corresponding to the medium phase state and the multi-parameter coupling error, and output the true physical thickness of the airport pavement media cover layer.

[0096] The parameter acquisition module is used to synchronously acquire multi-frequency impedance, surface temperature and environmental parameters of the pavement sensor. The system adopts a passive measurement principle and does not actively emit heat or high-frequency radio frequency waves, so as to realize the acquisition of reference signals in all-weather complex environments.

[0097] The salinity inversion and phase identification module has a built-in salinity-type calibration model. It extracts multi-dimensional feature vectors based on the acquired underlying electrical and thermodynamic data to achieve intelligent and accurate classification of the type of covering material.

[0098] The three-dimensional coupled adaptive calibration module has a built-in temperature-salinity-thickness calibration model. It calculates accurate salinity based on multi-dimensional coupling mapping, adaptively adjusts the impedance weights of differentiated characteristic frequencies and temperature compensation factors, and performs high-precision edge field contact physical thickness measurement.

[0099] The extracted feature parameters are input into a temperature-salinity coupled regression model. The regression model establishes a multidimensional coupled mapping by combining dielectric polarization mechanism and Debye relaxation characteristics, and outputs the corresponding basic salinity and temperature correction values. Based on the current pavement cover classification results, the temperature compensation factor and multi-frequency impedance weights are adaptively adjusted; and the dielectric constant and conductivity parameters in the thickness measurement benchmark are dynamically updated using the corrected accurate salt concentration; benchmark data are obtained under standard covers with known thickness and condition to determine the initial response parameters and edge electric field distribution constant of the sensor probe; and the temperature compensation factor is introduced in the calculation process to dynamically correct the inherent nonlinear electrical signal distortion caused by drastic ambient temperature drift and polarization of the measuring plate surface; a nonlinear conversion relationship between the electrical response signal and the actual medium thickness is established; based on the dielectric response principle, the impedance weight factor at characteristic frequency points is adaptively extracted and updated; and combined with dynamic weight allocation, the physical thickness of ice and water film is accurately calculated.

[0100] In summary, compared with the prior art, the technical solutions presented by this invention can achieve the following beneficial technical effects:

[0101] 1. Existing passive pavement sensors fail to maintain their preset static physical references when exposed to de-icing fluid dilution or drastic temperature fluctuations. This invention innovatively introduces a temperature-salinity multidimensional coupling model, capable of calculating the precise salt concentration of the pavement liquid film in real time and dynamically updating underlying physical parameters such as dielectric constant accordingly. This mechanism effectively filters out electromagnetic response interference under multi-field coupling environments, fundamentally solving the problems of temperature drift and salinity misjudgment that occur with traditional sensors when faced with changes in de-icing fluid concentration and drastic temperature changes.

[0102] 2. Addressing the industry pain point that traditional non-contact or basic capacitive sensors struggle to accurately quantify the thickness of accumulated water and ice, this invention abandons a single thickness measurement logic and innovatively constructs a differentiated frequency response mapping. Combined with a nonlinear temperature compensation factor specifically designed for water and ice layers, the system can adaptively adjust the penetration depth weights of high and low frequency edge electric fields, eliminating calculation errors caused by dielectric abrupt changes during the phase transition of ice / water mixtures, and achieving high-precision physical quantification of millimeter-level coverings.

[0103] 3. Unlike traditional methods that rely on single variables or human experience, this invention extracts impedance amplitude ratios, phase characteristic differences, and Debye relaxation parameters reflecting polar molecular properties at different frequencies to construct a deep multidimensional feature vector. By inputting a weighted feature fusion strategy into an improved multilayer perceptron (MLP) model, the system can adaptively and accurately distinguish easily confused physical states such as dryness, humidity, water accumulation, ice accumulation, and snow accumulation without polarization calibration, significantly reducing the false alarm rate caused by environmental noise.

[0104] The present invention also proposes a storage medium storing a computer program, which, when executed by a processor, implements a method for calibrating the thickness of the airport pavement medium cover layer.

[0105] In the above embodiments, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0106] Optionally, the processor executes the method steps described in the above embodiments based on the program code stored in the storage medium.

[0107] Specific examples in this embodiment can be found in the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0108] Obviously, those skilled in the art should understand that the modules or steps of the present disclosure described above can be implemented using general-purpose computing devices, which can be centralized on a single computing device or distributed on a network of multiple computing devices.

[0109] Alternatively, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computing device. Furthermore, in some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this disclosure is not limited to any particular combination of hardware and software.

[0110] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for calibrating the thickness of an airport pavement medium cover layer, characterized in that, include: Multi-frequency electrical parameters and environmental thermodynamic parameters are collected using medium sensors deployed on the airport pavement. The multi-frequency electrical parameters include preset multi-frequency impedance signals, and the environmental thermodynamic parameters include the real-time pavement temperature. The real-time salt concentration of the medium is obtained by inversion based on the multi-frequency electrical parameters. The phase type of the current medium covering layer of the pavement is identified by combining the multi-frequency electrical parameters and the real-time pavement temperature. The medium covering layer includes water film, de-icing film, ice accumulation, and snow accumulation. Configure corresponding multi-frequency combinations and impedance weights for different media phase states; The original thickness of the uncorrected dielectric overlay layer is initially calculated based on the preset multi-frequency impedance signal. Based on the multi-frequency electrical parameters, the real-time temperature of the pavement, the real-time salt concentration of the medium, the original thickness of the medium cover layer, the phase identification of the medium, and the Debye relaxation inherent parameters of various media, a three-dimensional coupled correction model of temperature-salinity-medium thickness is constructed. The nonlinear offset of the complex permittivity of the medium under different working conditions is calculated to obtain the multi-parameter coupling error. Based on the impedance weight corresponding to the phase state of the medium and the multi-parameter coupling error, the original thickness of the medium cover layer is dynamically corrected to eliminate the thickness measurement deviation caused by temperature drift, salinity interference, and phase state differences, and output the true physical thickness of the airport pavement medium cover layer.

2. The airport pavement medium cover layer thickness calibration method according to claim 1, characterized in that, The water film and the de-icing film use frequencies of 20kHz and 100kHz respectively, while the ice and snow use frequencies of 5kHz, 10kHz and 100kHz respectively.

3. The airport pavement medium cover layer thickness calibration method according to claim 2, characterized in that, Based on the Debye relaxation mechanism and the sensor edge electric field distribution constant, a thickness conversion relationship is established. Corresponding frequency points are matched for different medium phase states and impedance weighting factors are adaptively allocated. Independent nonlinear temperature compensation factors are configured for various phase states. After correcting the nonlinear distortion of electrical signals caused by plate polarization, the medium thickness is calculated.

4. The airport pavement medium cover layer thickness calibration method according to claim 1, characterized in that, The method for obtaining the real-time salt concentration of the medium based on the multi-frequency electrical parameters is as follows: substitute the multi-frequency impedance amplitude and phase into the pre-calibrated salinity response mapping curve, and solve to obtain the real-time salt concentration of the medium.

5. The airport pavement medium cover layer thickness calibration method according to claim 1, characterized in that, The phase state type of the medium covering layer is identified using a multilayer perceptron classification model, with the multi-frequency electrical parameters, the real-time pavement temperature, and the real-time salt concentration of the medium as the model input features.

6. The airport pavement medium cover layer thickness calibration method according to claim 1, characterized in that, The Debye relaxation intrinsic parameters corresponding to the various media include the static dielectric constant, the high-frequency limiting dielectric constant, and the dipole relaxation time.

7. The airport pavement medium cover layer thickness calibration method according to claim 1, characterized in that, The temperature-salinity-medium thickness three-dimensional coupled correction model uses the complex permittivity offset as an intermediate variable to establish a nonlinear correlation surface among temperature, salt concentration, and medium thickness.

8. The airport pavement medium cover layer thickness calibration method according to claim 1, characterized in that, After dynamically correcting the original thickness of the medium cover layer, the corrected actual physical thickness, corresponding medium phase state, and real-time operating parameters are synchronously stored in the local time series database.

9. A system for calibrating the thickness of an airport pavement medium cover layer, characterized in that, include: The parameter acquisition module is used to acquire multi-frequency electrical parameters and environmental thermodynamic parameters of the airport pavement. The multi-frequency electrical parameters include preset multi-frequency impedance signals, and the environmental thermodynamic parameters include the real-time pavement temperature. The salinity inversion and phase identification module is used to invert the real-time salt concentration of the medium based on the multi-frequency electrical parameters, and to identify the phase type of the current pavement medium covering layer by combining the multi-frequency electrical parameters and the real-time pavement temperature. The medium covering layer includes water film, de-icing film, ice accumulation, and snow accumulation. The frequency weight configuration and original thickness calculation module is used to configure the corresponding multi-frequency combination and impedance weight for different media phase states, and to preliminarily calculate the uncorrected original thickness of the dielectric cover layer based on the preset multi-frequency impedance signal. The three-dimensional coupled adaptive calibration module is used to combine the multi-frequency electrical parameters, the real-time pavement temperature, the real-time salt concentration of the medium, the original thickness of the medium cover layer, the medium phase identifier, and the Debye relaxation intrinsic parameters corresponding to various media to construct a three-dimensional coupled correction model of temperature-salinity-medium thickness, calculate the multi-parameter coupling error, and dynamically correct the original thickness of the medium cover layer according to the impedance weight corresponding to the medium phase and the multi-parameter coupling error, and output the true physical thickness of the airport pavement medium cover layer.

10. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a processor, implements the airport pavement medium cover thickness calibration method according to any one of claims 1 to 8.