A method for measuring ice thickness based on laser technology

Abstract

The present application relates to the technical field of optical thickness measurement, in particular to a kind of ice thickness measurement method based on laser technology, comprising the following steps: first and second wave band laser beam coaxial superposition lock initial phase difference generation calibration set, analysis beam table bottom interface reflection transmission and utilize detector to capture waveform, separate time of flight difference and compare double wave band propagation time delay to generate characteristic matrix, extract propagation power attenuation and combine dynamic refractive index to correct light speed to obtain geometric thickness result numerical value.In the present application, by constructing double wave band interface reflection analysis mechanism, the non-uniform beam tracking denoising ability is strengthened, the distortion interference caused by multipath scattering is effectively solved, the inherent wave band delay jitter is separated, and the signal attenuation caused by medium loss is reduced, combined with refractive compensation attenuation, the physical space solving accuracy is corrected, the bubble and inclination optical path error are eliminated, the detection purity is ensured, and the ice layer physical thickness reconstruction stability and monitoring precision are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of optical thickness measurement technology, and in particular to a method for measuring ice thickness based on laser technology. Background Technology

[0002] The field of optical thickness measurement technology involves a technical system that utilizes the optical properties of light, such as propagation, reflection, refraction, scattering, and interference, to determine the thickness, interlayer distance, interface position, and thickness distribution of a target object. Its core aspects include the emission mode of the measurement light source, the optical path configuration, the optical response of the surface and internal interface of the object being measured, the acquisition of returned light by the receiving device, the establishment of the measurement benchmark, the determination of the thickness conversion basis, and the measurement adaptation under different media conditions. Overall, it covers contact and non-contact measurement, single-point measurement and scanning measurement, static measurement and dynamic measurement, and widely involves the thickness determination of objects such as transparent bodies, translucent bodies, coated structures, liquid layers, and ice layers. Traditional ice thickness measurement methods refer to methods that address the technical issue of ice thickness by using a laser emitter to emit a laser beam onto the ice surface. The thickness is determined by analyzing the reflected echoes formed by the laser on the upper and lower surfaces of the ice, or at the ice-water interface and ice-base interface, combined with the propagation distance from the emission point to the reflection point, the laser incident angle, the refraction path, and the refractive index of the ice. A fixed bracket is used to install the laser and receiver, and a single distance measurement or point-by-point scanning is performed in a predetermined direction. The distance value of the upper surface of the ice is first obtained, and then the distance value of the lower interface of the ice is obtained. Subsequently, the ice thickness is calculated based on the two distance measurement results, the optical path difference relationship, or the trigonometric measurement relationship. In some applications, a calibration plate, a reference surface, an angle adjustment mechanism, a position movement mechanism, and a time interval recording step are also used to constrain the measurement process.

[0003] Traditional ice thickness measurement uses a fixed bracket to install transceiver equipment and scans the distance in a predetermined direction, either once or point by point. It relies on the optical path difference of the reflected signals from the two interfaces and simple spatial geometric relationships to calculate the thickness. Fixed single-band ranging is easily affected by internal bubbles and multipath scattering when facing complex, variable, and non-uniform media. This results in significant distortion deviations in the extracted echo time and cannot effectively cope with the energy attenuation and dynamic changes in refractive index of light during the penetration process. As a result, the measurement process is prone to accumulating errors, and the output geometric thickness deviates from the true physical form, leading to a decrease in overall monitoring accuracy. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a method for measuring ice thickness based on laser technology, comprising the following steps: To achieve the above objectives, the present invention adopts the following technical solution: a method for measuring ice thickness based on laser technology, comprising the following steps: S1: Obtain the first-band laser beam and the second-band laser beam, coaxially superimpose the two laser beams through the beam combiner assembly, adjust the spatial divergence angle and optical axis coincidence, lock the initial phase difference between the bands, and generate a dual-band laser coaxial calibration dataset. S2: Based on the dual-band laser coaxial calibration dataset, drive the laser to axially penetrate the ice surface, analyze the first specular reflection of the beam on the ice surface, and analyze the secondary reflection of the transmitted beam at the bottom of the ice layer to generate a set of ice layer dual-interface reflection characteristic parameters. S3: Call the ice layer dual interface reflection characteristic parameter set, use a photodetector to capture the double pulse waveform after photoelectric conversion, separate the optical flight time difference between the surface echo and the bottom echo, compare the propagation delay of the dual bands in the ice layer medium, and generate the ice layer interface echo time delay characteristic matrix. S4: Based on the echo delay feature matrix of the ice layer interface, extract the peak power attenuation of each band pulse after propagation inside the ice layer, correct the speed of light by combining the dynamic refractive index of the ice layer medium, calculate the real-time geometric distance through the physical correlation between the power attenuation model and the time delay, and generate the numerical value of the ice layer thickness measurement result.

[0005] As a further aspect of the present invention, the dual-band laser coaxial calibration dataset includes the composite beam center wavelength, spot centroid deviation, interference phase lock value, and initial divergence angle coefficient; the ice layer dual-interface reflection characteristic parameter set includes surface reflected light intensity distribution, bottom layer transmission loss, multipath scattering offset vector, and dual-interface spatial coordinate mapping; the ice layer interface echo time delay characteristic matrix includes the first band flight time, the second band flight time, surface-bottom layer pulse interval, and dispersion time delay jitter component; and the ice layer thickness measurement result values ​​include band-independent thickness calculation values, absorption spectrum attenuation rate, medium refractive index dynamic correction amount, and cosine tilt angle compensation value.

[0006] As a further aspect of the present invention, the steps for obtaining the dual-band laser coaxial calibration dataset are as follows: S101: Acquire the first-band laser beam and the second-band laser beam, adjust the transmission axis of the first laser and the reflection axis of the second laser to be orthogonally coincident, eliminate spatial optical axis deviation error, and generate an initial coaxial composite beam; S102: Based on the initial coaxial composite beam, a portion of the monitoring light is extracted by a beam splitter, and an image sensor is used to analyze the energy centroid coincidence and divergence angle matching degree of the two laser beams on the beam cross section to generate a spatial coupling coincidence matrix. S103: Call the spatial coupling overlap matrix to lock the polarization state and interference phase of the two laser beams, compensate for power fluctuations in the optical path transmission, output a composite beam, and generate a dual-band laser coaxial calibration dataset.

[0007] As a further aspect of the present invention, the steps of establishing the ice layer dual-interface reflection characteristic parameter set are as follows: S201: Based on the coaxial beam of the dual-band laser coaxial calibration dataset, control the laser emitter to aim the key deflection angle at the ice target, capture the energy distribution of the first interface reflected light at the air-ice interface through the image sensor, and generate the initial surface reflected beam vector. S202: Based on the initial reflected beam vector of the surface, track the propagation trajectory of the remaining transmitted beam inside the ice layer, quantify the secondary reflection and scattering process when the beam passes through the non-uniform medium and reaches the ice-water interface, and generate the bottom layer transmission and reflection light path distribution data. S203: Call the underlying transmission and reflection optical path distribution data, perform spatial filtering on the received dual-interface optical echo, filter out multipath scattered polarized light caused by bubbles or impurities inside the ice layer, and obtain the ice layer dual-interface reflection characteristic parameter set.

[0008] As a further aspect of the present invention, the step of constructing the ice layer interface echo delay feature matrix specifically includes: S301: Call the ice layer dual-interface reflection characteristic parameter set, focus the optical signal onto the photodetector target surface, and convert it into a dual-peak voltage pulse sequence through a high-speed sampling circuit to generate a photoelectric conversion pulse waveform set; S302: Call the photoelectric conversion pulse waveform set, use leading-edge phase detection or constant fractional discrimination technology to locate the triggering time of the surface echo pulse and the bottom echo pulse, calculate the absolute optical time of flight, and generate the propagation time of flight of the surface and bottom interface. S303: Based on the propagation flight time of the surface interface, compare the group velocity dispersion caused by the wavelength difference between the first band and the second band, separate the inherent time delay jitter components between the bands, and obtain the ice interface echo time delay feature matrix.

[0009] As a further aspect of the present invention, the steps for obtaining the numerical value of the ice layer thickness measurement result are specifically as follows: S401: Based on the echo delay characteristic matrix of the ice layer interface, compare the transmitted power and the received peak power, and combine the absorption cross section of the ice layer for different wavelengths to establish a power attenuation model and generate a band-related energy attenuation coefficient. S402: Based on the band-related energy attenuation coefficient, the refractive index constant of the ice layer medium is introduced, and dynamic refractive index compensation calculation is performed on the real-time propagation speed of the beam in the ice layer to generate the light speed correction amount after refractive index compensation. S403: Call the light speed correction amount after refractive index compensation, perform physical-level geometric calculations with the pure time delay data, deduct the cosine distortion error caused by the beam obliquely entering the ice surface, and generate the ice thickness measurement result value.

[0010] As a further aspect of the present invention, the pure time delay data refers to the stable time delay sequence formed by removing abnormal time delay values ​​in the ice layer interface echo time delay feature matrix according to a preset time delay deviation threshold, and performing mean filtering on the remaining time delay data, thus obtaining pure time delay data. The physical-level geometric solution refers to synchronously matching the refractive index-compensated light speed correction with the pure time delay data, constructing a propagation path length constraint relationship according to a preset optical path calculation rule, and performing spatial geometric solution in combination with the beam propagation direction parameters.

[0011] As a further aspect of the present invention, the method further includes step S5: S5: Call the measured ice thickness values, compare the thickness deviation of the independent measurements in the dual bands, quantify the medium dispersion effect, introduce the ice density physical model for volume scattering compensation, eliminate the optical path error caused by refraction, and obtain a detailed table of real-time monitoring and analysis of ice thickness.

[0012] As a further aspect of the present invention, the detailed table of real-time monitoring and analysis of ice layer thickness includes absolute optical physical thickness, dual-band dispersion correction, volume scattering compensation gain, and spatial distance calibration constant.

[0013] As a further aspect of the present invention, the steps of the real-time monitoring and analysis details table of ice layer thickness are as follows: S501: Call the multi-band measurement components in the ice layer thickness measurement result, extract the independent thickness values ​​calculated by the first band and the second band, identify the dispersion displacement caused by the non-uniformity of the ice layer refractive index through the difference operation, and generate the dual-band dispersion displacement difference value. S502: Call the dual-band dispersion shift difference, combine it with the scattering loss law of the beam inside the ice layer, introduce the volume optical physical correction model to perform secondary compensation of the geometric distance, and generate the medium coupling physical calibration quantity. S503: Based on the aforementioned medium coupling physical calibration value, it is superimposed on the spatial distance correction completed in the initial thickness measurement sequence, and the absolute physical measurement value eliminating optical distortion is output to obtain a detailed table of real-time monitoring and analysis of ice layer thickness.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by constructing a dual-band coaxial superposition and interface reflection feature analysis mechanism, the beam trajectory tracking and filtering noise reduction capabilities under heterogeneous media are enhanced, effectively solving the waveform distortion interference caused by multipath scattering, extracting the flight time of independent bands and separating time delay jitter, reducing signal attenuation caused by nonlinear physical loss between air and ice-water interface, introducing refractive index dynamic compensation and attenuation model to correct physical time delay, optimizing the physical-level spatial calculation accuracy of absolute geometric distance, eliminating optical path error caused by internal bubbles and tilt angle, enhancing adaptability to complex environments and ensuring the purity of detection data, achieving efficient fusion of multidimensional features of medium refractive index and absorption spectrum, and greatly improving the stability of physical thickness reconstruction and real-time monitoring accuracy. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0019] Please see Figure 1 This invention provides a method for measuring ice thickness based on laser technology, comprising the following steps: S1: Obtain the first-band laser beam and the second-band laser beam, coaxially superimpose the two laser beams through the beam combiner assembly, adjust the spatial divergence angle and optical axis coincidence, lock the initial phase difference between the bands, and generate a dual-band laser coaxial calibration dataset. S2: Based on the dual-band laser coaxial calibration dataset, drive the laser to be axially incident on the surface of the ice layer, analyze the first specular reflection of the beam on the surface of the ice layer, and analyze the secondary reflection of the transmitted beam at the bottom of the ice layer to generate a set of reflection characteristic parameters of the ice layer at the dual interface. S3: Call the ice layer dual-interface reflection characteristic parameter set, use a photodetector to capture the double pulse waveform after photoelectric conversion, separate the optical flight time difference between the surface echo and the bottom echo, compare the propagation delay of the two bands in the ice layer medium, and generate the ice layer interface echo time delay characteristic matrix. S4: Based on the echo delay feature matrix of the ice layer interface, extract the peak power attenuation of each band pulse after propagation inside the ice layer, combine the dynamic refractive index of the ice layer medium to correct the speed of light, and calculate the real-time geometric distance through the physical correlation between the power attenuation model and the time delay to generate the numerical value of the ice layer thickness measurement result. S5: Call the ice thickness measurement results, compare the thickness deviation of the independent dual-band measurements, quantify the medium dispersion effect, introduce the ice density physical model for volume scattering compensation, eliminate the optical path error caused by refraction, and obtain a detailed table of real-time monitoring and analysis of ice thickness.

[0020] The dual-band laser coaxial calibration dataset includes the composite beam center wavelength, spot centroid deviation, interference phase lock value, and initial divergence angle coefficient. The ice layer dual-interface reflection characteristic parameter set includes surface reflected light intensity distribution, bottom layer transmission loss, multipath scattering offset vector, and dual-interface spatial coordinate mapping. The ice layer interface echo time delay characteristic matrix includes the first band flight time, the second band flight time, surface-bottom layer pulse interval, and dispersion time delay jitter component. The ice layer thickness measurement results include band-independent thickness calculation values, absorption spectrum attenuation rate, medium refractive index dynamic correction, and cosine tilt angle compensation value. The ice layer thickness real-time monitoring and analysis details table includes absolute optical physical thickness, dual-band dispersion correction, volume scattering compensation gain, and spatial distance calibration constant.

[0021] Please see Figure 2 The specific steps for dual-band laser coaxial calibration dataset are as follows: S101: Acquire the first-band laser beam and the second-band laser beam, adjust the transmission axis of the first laser and the reflection axis of the second laser to be orthogonally coincident, eliminate spatial optical axis deviation error, and generate an initial coaxial composite beam; A continuous green laser with a wavelength of 532 nm is emitted as the first-band laser beam, and a pulsed near-infrared laser with a wavelength of 1064 nm is emitted as the second-band laser beam. Discrete power is captured at a sampling frequency of 1000 Hz within a 1-second time window. Global integration is performed on the data from 1000 sampling points, and the quotient is calculated by combining the total frequency. If the total power integral is 5000 mW, the initial reference power value for the first-band is calculated to be 5.0 mW. Similarly, the initial reference power value for the second-band is calculated to be 12.0 mW. The first-band laser beam is controlled to pass perpendicularly through the polarization-transmitting isolation layer at 0 degrees, and the second-band laser beam is controlled to be incident on the isolation layer at 45 degrees and undergo a 90-degree reflection deflection to achieve spatial collinearity. The coordinates of the energy distribution center of the two bands are extracted on a 1000 mm detection surface, and the absolute value of the actual spatial deviation error is derived by performing a square root operation on the difference between the horizontal and vertical coordinates. With the first band's x and y coordinates set to 100.50 mm and 150.20 mm, and the second band to 100.53 mm and 150.16 mm, the calculated absolute spatial deviation is 0.05 mm. A reference value of 0.02 mm for optical axis alignment is used for verification. This reference value is set based on the physical boundary of maintaining an energy coupling rate of over 95% in 50 consecutive tests. A deviation of 0.05 mm is greater than 0.02 mm, triggering reverse position compensation. The 0.03 mm error overscalar is extracted and multiplied by a spatial feedback scaling factor of 50 micrometers per millimeter, resulting in a 1.5 micrometer micro-motion adjustment command. This process is iterated until the spatial deviation error is reduced to 0.015 mm, generating an initial coaxial composite beam. Experimental results show an 82% improvement in alignment accuracy.

[0022] S102: Based on the initial coaxial composite beam, a portion of the monitoring light is extracted by the beam splitting element, and the energy centroid coincidence and divergence angle matching degree of the two laser beams on the beam cross section are analyzed by the image sensor to generate a spatial coupling coincidence matrix. Based on the initial coaxial composite beam, an energy splitting mechanism is introduced to strip 5% of the beam into the monitoring channel, obtaining a 1024x1024 pixel grayscale distribution matrix of the light spot. Fifty frames of background data under no-light conditions are extracted, and numerical integration and mean conversion are performed to establish a background reference value for each pixel. Differential noise reduction is applied to the original grayscale distribution matrix and the background reference value, with negative values ​​forced to zero. The horizontal energy moment distribution integral and the absolute energy total integral of the first band are calculated, and their quotient is used to derive the horizontal coordinate of the energy centroid of the first band. Similarly, the horizontal and vertical energy centroid coordinates of both bands are derived. The horizontal and vertical coordinates of the first band are set to 512.5 pixels and 512.0 pixels, and the second band to 513.0 pixels and 511.5 pixels. Euclidean distance measurement is performed on the difference between the two sets of coordinates to obtain a comprehensive centroid deviation distance of 0.707 pixels. The difference in light spot radius is measured at different distance points along the transmission path, and the actual spatial divergence angle is deduced based on the propagation distance. The absolute value of the difference in divergence angles between band 1 and band 2 was extracted. With band 1 defined as 1.50 mradians and band 2 as 1.55 mradians, the calculated divergence angle deviation was 0.05 mradians. The energy centroid coordinates of band 1 and band 2, the combined centroid deviation distance, and the divergence angle deviation were integrated into a multi-dimensional data array to construct a spatial coupling overlap matrix. Experimental results showed that the centroid localization noise resistance was improved by 85%.

[0023] S103: Call the spatial coupling overlap matrix to lock the polarization state and interference phase of the two laser beams, compensate for power fluctuations in the optical path transmission, output a composite beam, and generate a dual-band laser coaxial calibration dataset. The divergence angle deviation value of the spatial coupling overlap matrix is ​​extracted and compared with the allowable upper limit of 0.10 milliradians (mRA). A deviation value of 0.05 mRA meets the preconditions for physical interference, triggering the electro-optic phase feedback mechanism. The real-time interference fringe contrast value of the composite beam is captured and quantized with the ideal interference contrast reference value of 0.95. This reference value originates from the isothermal environment limit contrast measurement; assuming a real-time contrast of 0.85, the absolute difference of 0.10 is calculated. This difference is then product-coupled with the electro-optic crystal driving voltage coefficient of 200 volts per unit, outputting a 20-volt phase adjustment feedback voltage which is applied to the phase modulation circuit to lock the zero-phase-difference state. During the power fluctuation compensation stage, the transient total output power of the composite beam is sampled in real time and differentially quantized with the expected rated power of 17.0 milliwatts. If the sampled power is 16.5 milliwatts, a negative 0.5 milliwatt attenuation error value is derived. Activate the feedforward closed-loop proportional-integral control logic, combining a 2.0 proportional coefficient and a 0.5 mW / s integral coefficient, and perform parameter aggregation by summing the current error with the historical 0.2 mW error integral. Solve for the -1.0 mW proportional drive component and the updated integral component, and output a comprehensive drive current adjustment command to adjust the pump source injection intensity. Iterate until the absolute difference in dynamic power remains below 0.01 mW for 10 seconds, then synchronize and encapsulate the phase feedback voltage and transient output power timestamps to generate a dual-band laser coaxial calibration dataset.

[0024] Please see Figure 3 The specific steps for obtaining the ice layer dual-interface reflection feature parameter set are as follows: S201: Based on the coaxial beam of the dual-band laser coaxial calibration dataset, the laser emitter is controlled to aim the key deflection angle at the ice target. The energy distribution of the first interface reflected light at the air-ice interface is captured by the image sensor to generate the initial surface reflected beam vector. Based on a coaxial beam using a dual-band laser coaxial calibration dataset, the initial azimuth and elevation angle data are used, combined with the spatial deflection angle parameters transformed from the geographic coordinates of the ice layer under test. Differential calculations are performed to obtain the azimuth rotation deviation and elevation deviation, driving a multi-axis gimbal for positioning and tracking. When the beam reaches the interface between the air and ice layer, the optical receiving array captures the reflected signal and converts it into a 256x256 pixel input tensor. This tensor is then input into the first processing layer of a deep spatial perception network, where 32 3x3 scale weight matrices are used to perform sliding window product integration, combined with a threshold truncation function to eliminate negative backscattering noise. The output data enters a pooling downsampling network layer, where a 2x2 sliding window is used to extract local maxima, compressing the tensor dimension to 128x128. Entering the second processing layer, 64 5x5 scale weight matrices are used to perform global receptive field feature aggregation. After one-dimensional tiling, the results are fed into a mapping and aggregation layer containing 512 numerical nodes. Fully connected weighted multiplication and accumulation are performed, outputting a predicted total reflectance of 8%, a main lobe broadening of 2.5 milliradians, and a scattering deviation of 0.12 radians. These are compared with the environmental noise floor reflectance constant of 0.5% to obtain a 16-fold signal-to-noise ratio coefficient. After confirming the feature authenticity, multi-dimensional parameter stitching is performed to generate the initial surface reflected beam vector.

[0025] S202: Based on the initial reflected beam vector on the surface, the propagation trajectory of the remaining transmitted beam inside the ice layer is traced, the secondary reflection and scattering process of the beam when it passes through the non-uniform medium and reaches the ice-water interface is quantified, and the underlying transmitted and reflected light path distribution data is generated. The surface reflection loss parameter value was extracted and differentially discretized with the initial incident energy of 1000 microjoules to obtain the initial transmitted beam energy of 920 microjoules. In the trajectory tracking and quantization stage, the ice layer depth layer spacing was set to 0.1 meters. A 1.2% absorption attenuation constant and a 0.8% particle scattering attenuation constant per meter were used to obtain a 2.0% comprehensive attenuation coefficient per meter through numerical assembly. In the layer-by-layer micro-evolution calculation, the top incident energy, comprehensive attenuation coefficient, and layer spacing were multiplied for attenuation evaluation to derive the single-layer energy loss value, and differential stripping was performed to obtain the remaining energy. After 50 layers of recursive evolution calculations, the remaining energy in the deep layer attenuated to 480 microjoules, reaching the ice-water interface. The 15% interface reflection ratio constant calibrated by the standard seawater and freshwater freezing medium refractive index mutation model was retrieved, and the 480 microjoules were multiplied and fused with the 15% interface reflection ratio to calculate the absolute energy of the secondary echo emission as 72 microjoules. Substituting the 72 microjoules of secondary echo energy into the volumetric spatial scattering model, the hemispherical reflection space is divided into 72 sector-shaped solid angle regions. Using the Monte Carlo random walk algorithm, the echo energy components and time delays of each region are calculated. Multidimensional data mapping and orchestration are then performed to generate the underlying transmission and reflection optical path distribution data.

[0026] S203: Call the underlying transmission and reflection optical path distribution data, perform spatial filtering on the received dual-interface optical echo, filter out multipath scattered polarized light caused by bubbles or impurities inside the ice layer, and obtain the ice layer dual-interface reflection characteristic parameter set. The underlying transmission and reflection light path distribution data is used to introduce a deep spatial multi-level filtering classification network. The network input layer receives a 512x512 grid spatial spot intensity tensor, performs data deviation standardization, extracts the global intensity integral mean and standard deviation, and extracts the difference between the original pixel value and the mean, dividing by the standard deviation to achieve dimensionless processing. The data enters the first filtering hidden layer, where 64 3x3 regional feature sliding windows are used to perform multiplication and accumulation to locate high-frequency abrupt change points at impurity edges, and a nonlinear truncation function is used to clear dark spot redundancy. The data then enters the feature compression layer, where a 2x2 sliding step size is used to perform maximum downsampling and compression to a 256x256 dimension. In the second filtering hidden layer, 128 5x5 scale sliding windows are used to mine global polarization displacement features, which are then mapped to a global weight analysis layer containing 1000 parameter nodes, outputting bubble interference probability identification values. The model employs a cross-entropy error convergence function and an adaptive momentum correction update mechanism. The learning rate hyperparameter of 0.002 is established through exhaustive optimization based on 10,000 sets of measured data. Probability identification values ​​are extracted and compared with a multipath interference rejection benchmark threshold of 0.4 for size verification. If the probability value is 0.75, exceeding the 0.4 threshold, a data erasure operation is triggered; if it is 0.15, verification is approved. The effective signal peak extremes and flight time series spans are extracted and encapsulated to generate a set of ice layer dual-interface reflection feature parameters.

[0027] Please see Figure 4 The specific steps for obtaining the echo delay feature matrix of the ice layer interface are as follows: S301: Call the ice layer dual-interface reflection characteristic parameter set, focus the optical signal onto the photodetector target surface, convert it into a dual-peak voltage pulse sequence through a high-speed sampling circuit, and generate a photoelectric conversion pulse waveform set; By utilizing the dual-interface reflection characteristic parameter set of the ice layer and configuring the optical path and focal length parameters of the lens group based on the dual-interface reflection characteristics, the weak echo is compressed into a high-density 0.1 mm light spot and focused onto the effective target surface of a high-response avalanche photodiode. The carrier multiplication effect is excited, converting the discrete photon signal into a microampere-level fundamental photocurrent. The photocurrent is input into an extremely low-noise transimpedance amplification preamplifier network, and transimpedance product conversion logic is performed using a 1000-ohm conversion feedback impedance to amplify the microampere-level current and map it into a millivolt-level fundamental detection voltage fluctuation. An ultra-high frequency dynamic conversion circuit is introduced to perform digital discrete capture, with a sampling frequency calibrated to 2 billion times per second, achieving a time resolution accuracy on the order of 0.5 nanoseconds. To address the physical jump phenomenon between surface and subsurface echoes, a differential smoothing digital computation mechanism is used to eliminate high-frequency thermal noise. The current sampling point and five adjacent transient voltage values ​​are extracted, global integration is performed, and the average value is calculated using these five values, outputting a locally smoothed replacement voltage value. After undergoing high-frequency capture and smooth reconstruction operations, a continuous digital waveform array with a steep front end (double main lobe) is output in the digital domain. Multiple batches of probe waveforms are combined and stitched together to generate a set of photoelectric conversion pulse waveforms covering a comprehensive physical representation.

[0028] S302: Call the photoelectric conversion pulse waveform set, use the leading-edge phase detection or constant fractional discrimination technology to locate the trigger time of the surface echo pulse and the bottom echo pulse, calculate the absolute optical time of flight, and generate the propagation time of flight of the surface and bottom interface. The photoelectric conversion pulse waveform set is invoked to extract the single-cycle waveform array sequence. The system iterates through the first peak voltage reference in the surface echo region and the second peak voltage reference in the bottom echo region. A constant amplitude truncation of 0.5 is applied, calibrated based on an anti-white noise width extension experiment, corresponding to the extreme region of the rising edge rate of change. The first peak voltage reference is set to 4.6 volts. This is multiplied by the 0.5 constant amplitude ratio to extract the 2.3-volt absolute trigger level parameter for surface pulse phase detection. A point-by-point scan comparison is performed along the digital waveform axis. When the scan reaches 1250.00 nanoseconds (the 2500th sampling point), the recorded voltage is 2.28 volts, showing a negative polarity compared to 2.3 volts. The adjacent 2501st sampling point has a voltage of 2.35 volts, showing a positive polarity transition. Linear time difference interpolation calculations are initiated within the polarity alternation range to extract the difference between the 0.02 volt and 0.07 volt spans and perform proportional allocation derivation to obtain a 0.2857 time difference weight component. Combined with a 0.5 nanosecond reference time span, span product calculations are performed to derive a 0.1428 nanosecond micro-shift correction amount, which is then superimposed on a 1250.00 nanosecond reference, accurately locking the 1250.1428 nanosecond surface echo trigger timing. Similarly, the 2450.5428 nanosecond bottom echo trigger timing is obtained through derivation. Difference distance quantization is performed on the bottom and surface timings to obtain a 1200.4000 nanosecond absolute optical span delay. An assembly array is executed on multiple batches of time-consuming data to generate a time-of-flight record for the propagation from the bottom interface.

[0029] S303: Based on the propagation flight time of the surface interface, the group velocity dispersion caused by the wavelength difference between the first and second bands is compared to separate the inherent time delay jitter components between the bands, and the ice interface echo time delay feature matrix is ​​obtained. Based on the propagation flight time at the surface interface, multi-band independent transmission delay parameters are extracted from the time-lapse dataset. Dispersion difference isolation calculations caused by physical wavelength differences are introduced, utilizing the equivalent group refractive constants of 1.31 for band 1 and 1.29 for band 2. Assuming a reference ice thickness assessment scalar of 100 meters, and combining it with the vacuum light speed constant of 300,000,000 meters per second, the refractive constant, the reference scalar, and constant 2 are used to perform spatial return optical path product calculations and allocate the light speed quotient. Theoretical delay references of 873.33 nanoseconds for band 1 and 860.00 nanoseconds for band 2 are obtained. Differential isolation is performed on the two references to obtain a theoretical dispersion delay difference reference value of 13.33 nanoseconds. The actual flight delay of band 1 (1200.4000 nanoseconds) and band 2 (1182.2000 nanoseconds) were extracted, and differential quantization was performed to obtain the actual total time difference component of 18.2000 nanoseconds. A thickness magnification factor of 1.37 was applied, and this component was proportionally projected and multiplied with the theoretical dispersion delay difference reference value of 13.33 nanoseconds. The actual total time difference component was compared with the projected dispersion expectation value of 18.26 nanoseconds, and a micro-delay jitter of -0.06 nanoseconds was calculated. This micro-delay jitter was then removed, and the multi-dimensional pure delay parameters and timing channels were reconstructed into a digital array matrix, outputting the ice layer interface echo delay feature matrix.

[0030] Please see Figure 5 The specific steps for measuring the ice thickness are as follows: S401: Based on the echo delay characteristic matrix of the ice layer interface, the transmitted power and the received peak power are compared. Combined with the absorption cross section of the ice layer for different wavelengths, a power attenuation model is established to generate the band-related energy attenuation coefficient. The parameter configuration within the echo delay feature matrix of the ice layer interface is extracted. For Band 1, the initial transmit reference power setting of 5.0 mW and the receive peak absolute power of 0.001 mW are used. A quotient quantization calculation is performed at the transmit and receive ends to obtain a comprehensive energy transmission ratio of 0.0002. This ratio is then linearized using natural logarithmic extraction logic to calculate a negative logarithmic feature value of -8.517. The absorption cross-section physical parameter of 0.0015 unit cross-section parameter, a dedicated calibration for Band 1, is used. Combined with the estimated total bidirectional flight distance of 150 meters, a spatial product allocation calculation is performed to obtain a distance weight multiplier of 0.225. The absolute value of the negative logarithmic feature of 8.517 is used to derive a comprehensive attenuation characteristic of 37.853 by quotient derivation of the distance weight multiplier of 0.225. A multiple scattering compensation constant of 0.85 is introduced for coefficient scaling calculation to calculate the band-related energy attenuation coefficient of 32.17 per 100 meters for Band 1. Similarly, based on the 12.0 mW transmit value and 0.0005 mW receive value of the second band, and the absorption parameter of 0.02 per unit cross-section, its energy attenuation coefficient of 2.85 per 100 meters is calculated. Multi-band parameter alignment and layout are then performed to output a set of band-related energy attenuation coefficients.

[0031] S402: Based on the band-correlated energy attenuation coefficient, the refractive index constant of the ice medium is introduced, and dynamic refractive index compensation calculation is performed on the real-time propagation speed of the beam in the ice layer to generate the light speed correction amount after refractive index compensation. Based on the band-correlated energy attenuation coefficient, a deep-sensing dynamic refractive index extrapolation mechanism is constructed, invoking the scalar of the inherent refractive index constant of pure ice at 1.309. The energy attenuation coefficient of 32.17 per 100 meters for the first band, derived upstream, is used. Combining the negative correlation between the attenuation coefficient and porosity, a density mapping basis parameter of 0.0002 is introduced. The energy attenuation coefficient and the density basis parameter are multiplied and transformed using a fluctuation conversion operation to obtain an additional refractive index increment of 0.00643. The solidified refractive index constant and the calculated refractive index increment are then superimposed and calculated to establish an actual equivalent working refractive index of 1.31543 under specific attenuation conditions. A vacuum absolute speed of light constant of 300,000,000 meters per second is extracted as a reference, and a quotient rate reduction operation is performed between the speed of light constant reference and the actual equivalent working refractive index. The calculation yields the true physical phase velocity of light waves passing through a high-density ice layer as 228062283 meters per second. By using the cascaded calculation logic from environmental attenuation characteristics to optical parameters, the constant refraction calculation mode is replaced, and a set of characteristic parameters for light speed correction after high-fitness refractive index compensation is output.

[0032] S403: Call the light speed correction amount after refractive index compensation, perform physical-level geometric calculations with the pure time delay data, subtract the cosine distortion error caused by the beam obliquely entering the ice surface, and generate the ice thickness measurement result value. Pure delay data refers to the stable delay sequence formed after removing abnormal delay values ​​in the ice layer interface echo delay feature matrix according to a preset delay deviation threshold, and then performing mean filtering on the remaining delay data. Physical-level geometric solution refers to synchronously matching the refractive index-compensated light speed correction with the pure time delay data, constructing a propagation path length constraint relationship based on preset optical path calculation rules, and performing spatial geometric solution in combination with beam propagation direction parameters; The refractive index-compensated light speed correction is used to filter out abnormal noise in the delay records within the ice interface echo delay feature matrix. A digital filtering algorithm based on a five-point sliding integral and the mean ratio is then applied to generate a steady-state pure delay data sequence. The refractive index-compensated light speed correction of 228062283 meters per second is extracted and multiplied by the 1200.40 nanosecond single-pass bidirectional pure delay data to obtain a spatial span quantization, resulting in a transmission propagation round-trip straight optical path of 273.7659 meters. This path is then isolated by equal division with a constant of 2, stripping away multipath paths and obtaining a unidirectional beam straight-line penetration distance of 136.8829 meters. The current physical deflection incident angle of the payload's forward-looking transmitter is extracted as 15.0 degrees. The corresponding 0.9659 trigonometric function cosine projection scalar is then used to perform a spatial vector projection mapping product between the oblique straight-line penetration distance and the cosine projection scalar. After eliminating the spatial geometric deformation and cosine stretching errors introduced by the oblique view, the final absolute vertical geometric depth of 132.2152 meters was obtained by deduction and calculation. The cyclic batch processing operation was performed on the detection node cluster to output the ice thickness measurement result value fused with the three-dimensional latitude and longitude reference system.

[0033] Please see Figure 6 The specific steps for real-time monitoring and analysis of ice layer thickness details are as follows: S501: Call the multi-band measurement components in the ice thickness measurement results, extract the independent thickness values ​​calculated from the first and second bands, identify the dispersion shift caused by the non-uniformity of the ice refractive index through the difference operation, and generate the dual-band dispersion shift difference value. The multi-band measurement components from the ice thickness measurement results are used to perform a systematic sampling and separation operation on the converged multi-band thickness calculation array. The first absolute depth feature value of 132.2152 meters, derived using the 532 nm wavelength first band channel, is extracted. The second absolute depth feature value of 132.1810 meters, derived using the 1064 nm wavelength second band channel, is also extracted from the same timestamp. To address the differential dispersion hysteresis effect caused by imperfect ice crystal structures, subtraction differential quantization is directly performed on the two sets of independent thickness measurements. The relative spatial hysteresis difference is analyzed to be 0.0342 meters. The derived spatial difference is then subjected to absolute value pure quantization logic to remove interference factors from positive and negative polarities in subsequent vector calculations, ensuring that the displacement deviation is converted into a pure spatial scalar. The quantized value of 0.0342 meters directly characterizes the degree of spatiotemporal misalignment mapping of the optical band within the medium layer. By traversing the entire measurement trajectory sequence point by point and performing the same operation, a dual-band dispersion displacement difference value reflecting the microstructural distortion of the entire domain is generated.

[0034] S502: Call the dual-band dispersion shift difference, combine it with the scattering loss law of the beam inside the ice layer, introduce the volume optical physical correction model to perform secondary compensation of the geometric distance, and generate the medium coupling physical calibration quantity. The volume scattering correction weighting constant scalar, representing the intensity of volume scattering during multiple diffuse reflection of light waves, is extracted from the dual-band dispersion displacement difference. This constant, rigorously calibrated as an equivalent spatial expansion coefficient of 0.35, is verified through inversion of hundreds of core ice core samples using an integral sphere. The 0.0342-meter dual-band dispersion displacement difference value generated from the upstream solution is then extracted. A spatial distribution product projection operation is performed on the volume scattering correction weighting constant scalar and the dispersion displacement difference, spanning the dimension of the simple band difference, to derive a preliminary compensation length vector of 0.01197 meters. The extreme value constant parameter of 0.005 meters, used to evaluate the additional influence factors of ice aging and impurity deposition, is then merged and superimposed with this aging attenuation extreme value parameter. The resulting solution generates a composite displacement micro-value of 0.01697 meters. This value accurately characterizes the virtual optical path elongation effect caused by refractive index dispersion misalignment and multipath volume diffuse reflection. It performs traversal calculations and normalization and storage on the assembly array, and finally outputs a medium coupling physical calibration quantity with deep correction capability.

[0035] S503: Based on the physical calibration quantity of the medium coupling, it is superimposed on the initial thickness measurement sequence to complete the spatial distance correction, output the absolute physical measurement value to eliminate optical distortion, and obtain the detailed table of real-time monitoring and analysis of ice layer thickness; Based on the medium-coupled physical calibration quantity, the depth error fusion correction operation logic is initiated. For a specific coordinate node, the original first-band channel is extracted to deduce the initial absolute reference thickness result of 132.2152 meters. Simultaneously, the 0.01697-meter medium-coupled physical calibration quantity parameter associated with this geographic mapping point is extracted. To remove the pseudo-range caused by the virtual extension of light wave reflection and scattering inside the crystal, the initial reference thickness and the physical calibration quantity are subjected to inverse difference elimination operation. The deduction and calculation finally obtain the core true thickness value of 132.1982 meters, achieving the bottom-level peeling off of microwave multipath effect. The full-time and spatial domain detection trajectory data array is subjected to cyclic correction calculation, and the core true thickness value group after physical correction and the basic mapping marker matrix are compiled. The bottom-level source parameter set such as multi-band delay reference parameter and dynamic refractive index offset record are incorporated, and the structured layout operation is performed in accordance with the spatial topology structure specification to output a real-time monitoring detail table of ice layer thickness with highly fused latitude and longitude information and the true value of absolute distance measurement after eliminating dispersion distortion.

[0036] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the protection scope defined by the technical solution of the present invention.

Claims

1. A method for measuring ice thickness based on laser technology, characterized in that, Includes the following steps: S1: Obtain the first-band laser beam and the second-band laser beam, coaxially superimpose the two laser beams through the beam combiner assembly, adjust the spatial divergence angle and optical axis coincidence, lock the initial phase difference between the bands, and generate a dual-band laser coaxial calibration dataset. The specific steps for generating the dual-band laser coaxial calibration dataset are as follows: S101: Acquire the first-band laser beam and the second-band laser beam, adjust the transmission axis of the first laser and the reflection axis of the second laser to be orthogonally coincident, eliminate spatial optical axis deviation error, and generate an initial coaxial composite beam; S102: Based on the initial coaxial composite beam, a portion of the monitoring light is extracted by a beam splitter, and an image sensor is used to analyze the energy centroid coincidence and divergence angle matching degree of the two laser beams on the beam cross section to generate a spatial coupling coincidence matrix. S103: Call the spatial coupling overlap matrix to lock the polarization state and interference phase of the two laser beams, compensate for power fluctuations in the optical path transmission, output a composite beam, and generate a dual-band laser coaxial calibration dataset. S2: Based on the dual-band laser coaxial calibration dataset, drive the laser to axially penetrate the ice surface, analyze the first specular reflection of the beam on the ice surface, and analyze the secondary reflection of the transmitted beam at the bottom of the ice layer to generate a set of ice layer dual-interface reflection characteristic parameters. S3: Call the ice layer dual interface reflection characteristic parameter set, use a photodetector to capture the double pulse waveform after photoelectric conversion, separate the optical flight time difference between the surface echo and the bottom echo, compare the propagation delay of the dual bands in the ice layer medium, and generate the ice layer interface echo time delay characteristic matrix. S4: Based on the echo delay feature matrix of the ice layer interface, extract the peak power attenuation of each band pulse after propagation inside the ice layer, correct the speed of light by combining the dynamic refractive index of the ice layer medium, calculate the real-time geometric distance through the physical correlation between the power attenuation model and the time delay, and generate the numerical value of the ice layer thickness measurement result.

2. The ice thickness measurement method based on laser technology according to claim 1, characterized in that, The dual-band laser coaxial calibration dataset includes the composite beam center wavelength, spot centroid deviation, interference phase lock value, and initial divergence angle coefficient. The ice layer dual-interface reflection characteristic parameter set includes surface reflected light intensity distribution, bottom layer transmission loss, multipath scattering offset vector, and dual-interface spatial coordinate mapping. The ice layer interface echo time delay characteristic matrix includes the first band flight time, the second band flight time, surface-bottom layer pulse interval, and dispersion time delay jitter component. The ice layer thickness measurement results include band-independent thickness calculation values, absorption spectrum attenuation rate, medium refractive index dynamic correction, and cosine tilt angle compensation value.

3. The method for measuring ice thickness based on laser technology according to claim 1, characterized in that, The specific steps for generating the ice layer dual-interface reflection feature parameter set are as follows: S201: Based on the coaxial beam of the dual-band laser coaxial calibration dataset, control the laser emitter to aim the key deflection angle at the ice target, capture the energy distribution of the first interface reflected light at the air-ice interface through the image sensor, and generate the initial surface reflected beam vector. S202: Based on the initial reflected beam vector of the surface, track the propagation trajectory of the remaining transmitted beam inside the ice layer, quantify the secondary reflection and scattering process when the beam passes through the non-uniform medium and reaches the ice-water interface, and generate the bottom layer transmission and reflection light path distribution data. S203: Call the underlying transmission and reflection optical path distribution data, perform spatial filtering on the received dual-interface optical echo, filter out multipath scattered polarized light caused by bubbles or impurities inside the ice layer, and generate a set of ice layer dual-interface reflection characteristic parameters.

4. The ice thickness measurement method based on laser technology according to claim 3, characterized in that, The specific steps for generating the ice layer interface echo delay feature matrix are as follows: S301: Call the ice layer dual-interface reflection characteristic parameter set, focus the optical signal onto the photodetector target surface, and convert it into a dual-peak voltage pulse sequence through a high-speed sampling circuit to generate a photoelectric conversion pulse waveform set; S302: Call the photoelectric conversion pulse waveform set, use leading-edge phase detection or constant fractional discrimination technology to locate the triggering time of the surface echo pulse and the bottom echo pulse, calculate the absolute optical time of flight, and generate the propagation time of flight of the surface and bottom interface. S303: Based on the propagation flight time of the surface interface, compare the group velocity dispersion caused by the wavelength difference between the first band and the second band, separate the inherent time delay jitter components between the bands, and generate the ice layer interface echo time delay feature matrix.

5. The method for measuring ice thickness based on laser technology according to claim 4, characterized in that, The specific steps for generating the numerical results of the ice layer thickness measurement are as follows: S401: Based on the echo delay characteristic matrix of the ice layer interface, compare the transmitted power and the received peak power, and combine the absorption cross section of the ice layer for different wavelengths to establish a power attenuation model and generate a band-related energy attenuation coefficient. S402: Based on the band-related energy attenuation coefficient, the refractive index constant of the ice medium is introduced, and dynamic refractive index compensation calculation is performed on the real-time propagation speed of the beam in the ice layer to generate the light speed correction amount after refractive index compensation. S403: Call the light speed correction amount after refractive index compensation, perform physical-level geometric calculations with the pure time delay data, deduct the cosine distortion error caused by the beam obliquely entering the ice surface, and generate the ice thickness measurement result value.

6. The method for measuring ice thickness based on laser technology according to claim 5, characterized in that, The pure time delay data refers to the stable time delay sequence formed by removing abnormal time delay values ​​in the ice layer interface echo time delay feature matrix according to a preset time delay deviation threshold, and then performing mean filtering on the remaining time delay data, thus obtaining pure time delay data. The physical-level geometric solution refers to synchronously matching the refractive index-compensated light speed correction with the pure time delay data, constructing a propagation path length constraint relationship according to a preset optical path calculation rule, and performing spatial geometric solution in combination with the beam propagation direction parameters.

7. The method for measuring ice thickness based on laser technology according to claim 1, characterized in that, The method also includes step S5: S5: Call the measured ice thickness values, compare the thickness deviation of the independent measurements in the dual bands, quantify the medium dispersion effect, introduce the ice density physical model for volume scattering compensation, eliminate the optical path error caused by refraction, and obtain a detailed table of real-time monitoring and analysis of ice thickness.

8. The method for measuring ice thickness based on laser technology according to claim 7, characterized in that, The detailed table of real-time monitoring and analysis of ice thickness includes absolute optical physical thickness, dual-band dispersion correction, volume scattering compensation gain, and spatial distance calibration constant.

9. The method for measuring ice thickness based on laser technology according to claim 7, characterized in that, The specific steps for obtaining the detailed table of real-time monitoring and analysis of ice layer thickness are as follows: S501: Call the multi-band measurement components in the ice layer thickness measurement result, extract the independent thickness values ​​calculated by the first band and the second band, identify the dispersion displacement caused by the non-uniformity of the ice layer refractive index through the difference operation, and generate the dual-band dispersion displacement difference value. S502: Call the dual-band dispersion shift difference, combine it with the scattering loss law of the beam inside the ice layer, introduce the volume optical physical correction model to perform secondary compensation of the geometric distance, and generate the medium coupling physical calibration quantity. S503: Based on the aforementioned medium coupling physical calibration value, it is superimposed on the spatial distance correction completed in the initial thickness measurement sequence, and the absolute physical measurement value eliminating optical distortion is output to obtain a detailed table of real-time monitoring and analysis of ice layer thickness.

Images