Method for obtaining data of seawater at different depths based on salinity sensor

Abstract

The present application relates to the field of marine science and technology, and specifically relates to a method for obtaining data of seawater at different depths based on a salinity sensor, comprising the following steps: S1: collecting salinity data of seawater at different depths in real time; S2: preprocessing the collected salinity data; S3: constructing a depth-salinity mathematical model by using a Gaussian process regression algorithm; S4: collecting environmental data in real time, and dynamically correcting parameters of the depth-salinity model according to the environmental data; and S5: inputting the preprocessed salinity data in S2 into the dynamically corrected depth-salinity mathematical model, and calculating corresponding seawater depth information. Through the Gaussian process regression algorithm and the dynamic correction technology, the present application constructs a nonlinear mathematical model of depth and salinity, and combines data preprocessing and inverse calculation to realize accurate prediction and dynamic adaptation of the relationship between salinity and depth in a complex marine environment.

Description

Technical Field

[0001] This invention relates to the field of marine science and technology, and in particular to a method for acquiring data of different depths of seawater based on a salinity sensor. Background Technology

[0002] With the increasing demand for marine resource development and environmental monitoring, accurately acquiring salinity data at different ocean depths has become an important technical challenge. Salinity data not only reflects the chemical characteristics of ocean water but also provides crucial support for marine ecosystem research, ocean current analysis, and climate change prediction. However, due to the complexity and dynamism of the marine environment, salinity distribution at different depths is affected by various factors, including water temperature, ocean current velocity, and seasonal variations. The acquisition of salinity data and depth correlation analysis present significant technical challenges. Traditional salinity measurement methods typically rely on single-point sampling or mechanical bathymetry equipment. These methods have significant limitations in terms of coverage, real-time performance, and data accuracy, making it difficult to meet the monitoring needs of complex marine environments.

[0003] In existing technologies, research on the depth-salinity relationship mainly focuses on empirical formulas or simple linear regression models. These methods cannot effectively capture the nonlinear relationship between depth and salinity. In addition, the dynamic changes in salinity under different marine environments have not been fully considered, resulting in poor adaptability of the prediction results. Especially in real-time changing marine environments, the influence of environmental factors such as water temperature and ocean current velocity on salinity distribution is often ignored, which limits the practical application of the model. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides a method for acquiring data of seawater at different depths based on a salinity sensor.

[0005] A method for acquiring data at different depths of seawater based on a salinity sensor includes the following steps:

[0006] S1: Deploy salinity sensors at different depths in the target ocean to collect real-time salinity data of seawater at different depths;

[0007] S2: Preprocess the collected salinity data, including noise removal, outlier removal, and missing data interpolation;

[0008] S3: Based on long-term historical salinity data, a depth-salinity mathematical model is constructed using the Gaussian process regression algorithm;

[0009] S4: Real-time collection of environmental data, including water temperature and ocean current velocity, and dynamic correction of the parameters of the depth-salinity model based on the environmental data, adjusting the kernel function and related hyperparameters in the mathematical model of depth and salinity to adapt to the real-time changing marine environment;

[0010] S5: Input the preprocessed salinity data from S2 into the dynamically corrected depth-salinity mathematical model to calculate the corresponding seawater depth information.

[0011] Optionally, S1 specifically includes:

[0012] S11: Based on the water depth distribution of the target sea area, determine the sampling depth location, specifically selecting several depth points in the surface layer (0-50 meters), the middle layer (50-200 meters), and the deep layer (below 200 meters).

[0013] S12: Deploy salinity sensors at each selected depth point, with each salinity sensor secured at the corresponding depth by a mooring cable or anchoring device;

[0014] S13: Utilize the depth gauge built into the salinity sensor to monitor and record its depth information in real time, ensuring that the collected salinity data matches the corresponding depth one by one.

[0015] Optionally, S2 specifically includes:

[0016] S21: Noise removal processing is performed on the collected salinity data. A bandpass filter is used to filter the salinity signal to remove high-frequency noise and low-frequency interference.

[0017] S22: Remove outliers from salinity data. Use the Z-Score algorithm to detect the salinity value of each data point. If the Z value of a data point exceeds the set threshold, the data point is considered an outlier and is removed.

[0018] S23: Perform interpolation to complete missing data. Use linear interpolation or spline interpolation to fill in the missing salinity data in the time series to ensure data continuity and consistency.

[0019] Optionally, S3 specifically includes:

[0020] S31: Obtain long-term historical salinity data for the target sea area, specifically including salinity observations at different times and depths, covering a period of at least two years;

[0021] S32: Normalize historical salinity data to bring salinity data at different depths into a standardized range. Specifically, use the minimum-maximum normalization method to map salinity values ​​to the interval [0, 1] to eliminate the magnitude differences between data at different depths and during observation periods.

[0022] S33: Based on the historical salinity data after normalization in S32, use the depth of each data record as the input variable and salinity as the output variable to construct a depth-salinity input-output pair to form a training dataset.

[0023] S34: Based on the training dataset formed in S33, the depth-salinity mathematical model is trained using the Gaussian process regression algorithm. Specifically, the radial basis function is selected as the kernel function, and the hyperparameters of the kernel function are optimized by the maximum likelihood estimation method, so that the depth-salinity mathematical model can capture the nonlinear relationship between depth and salinity.

[0024] Optionally, S33 specifically includes:

[0025] S331: Extract each data record from the historical salinity data after normalization in S32, including depth. and normalized salinity value , where i represents the data sequence number;

[0026] S332: Depth As the input variable of the model, the input vector X is constructed, and its expression is: Where n is the total number of data records, This represents the depth value of each record;

[0027] S333: Normalized salinity value As the output variable of the model, the output vector Y is constructed, and its expression is: ,in, This represents the salinity value corresponding to each record;

[0028] S334: Pair the input vector X and the output vector Y to form the training dataset T, which is expressed as follows: ;

[0029] S335: To improve the coverage of the training dataset, the training data is augmented; specifically, a linear interpolation method is used to fill in missing data within a continuous depth range.

[0030] Optionally, S34 specifically includes:

[0031] S341: Initialize the Gaussian process regression model, selecting the radial basis function as the kernel function to measure the similarity between different depths in the training data;

[0032] S342: Based on the training dataset, calculate the similarity of each pair of depth data to generate a covariance matrix. Each element in the covariance matrix represents the correlation between two depth locations.

[0033] S343: Calculate the predicted salinity value at any test depth based on the covariance matrix and the salinity value in the training data, and obtain the predicted value corresponding to the test depth through the similarity between the training depth and the test depth and the observation noise parameter.

[0034] S344: The similarity parameter and observation noise parameter of the kernel function are optimized. The optimization process is based on maximizing the probability value of the training data under the model to ensure that the model can best fit the training data, thereby obtaining the depth-salinity mathematical model.

[0035] Optionally, S4 specifically includes:

[0036] S41: Real-time data on water temperature and ocean current velocity in the target area are collected through deployed water temperature sensors and ocean current velocity sensors; the water temperature sensors are used to accurately monitor changes in seawater temperature; the ocean current velocity sensors are used to measure the flow velocity at the ocean surface or at different depths; at the same time, the collected data is preprocessed to remove noise and complete missing data to ensure data integrity.

[0037] S42: Based on preprocessed environmental data, analyze the effects of water temperature and ocean current velocity on salinity changes, and determine the relationship between environmental data and the depth-salinity model; where water temperature affects the solubility of salinity, and ocean current velocity affects the mixing and distribution of salinity;

[0038] S43: Based on the analysis results of S42, adjust the kernel function and related hyperparameters in the depth-salinity model; specifically, adjust the temperature factor in the kernel function according to water temperature changes to optimize the model's response at different water temperatures; and dynamically adjust the velocity factor in the kernel function according to changes in ocean current velocity.

[0039] Optionally, S42 specifically includes:

[0040] S421: Extract time-series data of water temperature T and ocean current velocity v based on preprocessed environmental data;

[0041] S422: Analyze the effect of water temperature T on salinity solubility, describing the relationship between water temperature changes and salinity through salinity solubility. The formula is: ,in, This indicates the effect of water temperature on salinity solubility. Indicates standard water temperature The salinity value at that time Temperature-related factors;

[0042] S423: Analyze the influence of ocean current velocity v on salinity mixing and distribution. Using a salinity diffusion model from fluid dynamics, describe the effect of ocean current velocity changes on the salinity gradient. The expression is: ,in, This represents the rate of change in salinity distribution caused by ocean current velocity. Ocean current influencing factors This represents the depth gradient of salinity.

[0043] S424: Integrate the effects of water temperature and ocean current velocity into a comprehensive model, constructing a comprehensive expression for the relationship between environmental data and salinity changes: ,in, This represents the salinity value at a specified depth d, influenced by the combined effects of water temperature and ocean current velocity.

[0044] Optionally, S43 specifically includes:

[0045] S431: Based on the comprehensive relationship expression of salinity changes using environmental data, extract the influence of environmental factors on the salinity model, specifically including water temperature factors. and flow rate factor ;

[0046] S432: Define the kernel function of the Gaussian process regression model and include the temperature factor. and flow rate factor As a hyperparameter of the Gaussian process kernel function;

[0047] S433: Based on the comprehensive relationship expression of salinity changes according to environmental data, dynamically adjust the hyperparameters in the kernel function; specifically, adjust the temperature factor through an optimization algorithm that maximizes the log-likelihood function. and flow rate factor To minimize prediction error;

[0048] S434: The optimal temperature factor is obtained by maximizing the log-likelihood function. and flow rate factor The optimized parameters are then fed back into the Gaussian process regression model to dynamically correct the depth-salinity mathematical model.

[0049] Optionally, S5 specifically includes:

[0050] S51: Input the preprocessed salinity data from S2 into the dynamically corrected depth-salinity mathematical model;

[0051] S52: Utilize the inverse calculation function of the depth-salinity mathematical model to calculate the corresponding seawater depth based on the salinity value;

[0052] S53: Perform confidence interval analysis on the calculation results. Quantify the uncertainty of the calculation results through the model's covariance matrix, and generate predicted values ​​and confidence intervals for each depth to ensure the reliability of the results.

[0053] S54: Organize all predicted depth results into a depth set, match it with the actual salinity data, and output the final salinity-depth relationship table.

[0054] The beneficial effects of this invention are:

[0055] This invention, by combining Gaussian process regression algorithm with dynamic correction mechanism, accurately constructs a nonlinear mathematical model between depth and salinity. It can comprehensively process salinity data at different depths as well as environmental factors such as real-time collected water temperature and ocean current velocity. By using Gaussian process regression algorithm to model historical salinity data and dynamically adjusting the kernel function and hyperparameters of the model under real-time environmental changes, it achieves high-precision prediction of salinity distribution, effectively solving the problem that traditional models cannot adapt to complex marine environmental changes, and significantly improving the prediction accuracy and adaptability of the relationship between salinity and depth.

[0056] This invention achieves high-precision derivation of seawater depth information from salinity data through data preprocessing, depth inverse calculation, and confidence interval analysis. The preprocessing technology ensures the integrity and consistency of the data, while the inverse calculation method, combined with a dynamically adjusted mathematical model, can accurately calculate the seawater depth information corresponding to salinity data, providing reliable technical support for marine ecological research, climate change analysis, and other related fields. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a schematic diagram of a method for acquiring data at different depths of seawater according to an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of a method for dynamically correcting parameters according to an embodiment of the present invention. Detailed Implementation

[0060] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0061] like Figures 1-2 As shown, the method for acquiring data at different depths of seawater based on a salinity sensor includes the following steps:

[0062] S1: Deploy salinity sensors at different depths in the target ocean to collect real-time salinity data of seawater at different depths;

[0063] S2: Preprocess the collected salinity data, including noise removal, outlier removal, and missing data interpolation, to make the data suitable for further analysis;

[0064] S3: Based on long-term historical salinity data, a depth-salinity mathematical model is constructed using the Gaussian process regression (GPR) algorithm. The model models the complex nonlinear relationship between salinity and depth by selecting an appropriate kernel function, which can effectively adapt to the hydrological characteristics and seasonal changes of different regions.

[0065] S4: Real-time collection of environmental data, including water temperature and ocean current velocity, and dynamic correction of the parameters of the depth-salinity model based on the environmental data, adjusting the kernel function and related hyperparameters in the mathematical model of depth and salinity to adapt to the real-time changing marine environment;

[0066] S5: Input the preprocessed salinity data from S2 into the dynamically corrected depth-salinity mathematical model to calculate the corresponding seawater depth information.

[0067] S1 specifically includes:

[0068] S11: Based on the water depth distribution of the target sea area, determine the sampling depth location, specifically selecting several depth points in the surface layer (0-50 meters), the middle layer (50-200 meters), and the deep layer (below 200 meters).

[0069] S12: Deploy salinity sensors at each selected depth point. Each salinity sensor is fixed at the corresponding depth by a mooring cable or anchoring device to ensure its stability and measurement accuracy.

[0070] S13: Utilize the depth gauge built into the salinity sensor to monitor and record its depth information in real time, ensuring that the collected salinity data matches the corresponding depth; through clear sampling depth selection and sensor deployment methods, ensure that accurate salinity data is obtained at different seawater depths, providing a reliable foundation for subsequent data analysis and model building.

[0071] The preprocessing of the collected salinity data in S2 specifically includes:

[0072] S21: The collected salinity data is subjected to noise removal processing. A bandpass filter is used to filter the salinity signal to remove high-frequency noise and low-frequency interference, ensuring the smoothness and effectiveness of the data. The cutoff frequency of the filter is set according to the fluctuation characteristics of the target sea area.

[0073] S22: Outlier removal for salinity data. The Z-Score algorithm is used to detect the salinity value of each data point. If the Z value of a data point exceeds the set threshold, the data point is considered an outlier and is removed. The threshold is set to 3 to ensure that most abnormal fluctuations are effectively identified and eliminated.

[0074] S23: Imput missing data by interpolation. Linear interpolation or spline interpolation is used to fill in the missing salinity data in the time series to ensure data continuity and consistency. The specific interpolation method is selected according to the location of the missing data and the pattern of adjacent data. Through preprocessing steps such as noise removal, outlier removal, and missing data interpolation, the quality of salinity data can be significantly improved, unnecessary interference information can be removed, and the accuracy and reliability of subsequent data analysis and model training can be ensured.

[0075] S3 specifically includes:

[0076] S31: Obtain long-term historical salinity data for the target sea area, specifically including salinity observations at different times and depths, covering a period of at least two years, to ensure that the data reflects the seasonal and hydrological changes of the sea area.

[0077] S32: Normalize historical salinity data to bring salinity data at different depths into a standardized range. Specifically, use the minimum-maximum normalization method to map salinity values ​​to the interval [0, 1] to eliminate the magnitude differences between data at different depths and during observation periods.

[0078] S33: Based on the historical salinity data after normalization in S32, use the depth of each data record as the input variable and salinity as the output variable to construct a depth-salinity input-output pair to form a training dataset; ensure that the input and output data can correspond accurately to facilitate model fitting;

[0079] S34: Based on the training dataset formed in S33, a depth-salinity mathematical model is trained using the Gaussian process regression algorithm. Specifically, the radial basis function is selected as the kernel function, and the hyperparameters of the kernel function are optimized using the maximum likelihood estimation method, enabling the depth-salinity mathematical model to capture the nonlinear relationship between depth and salinity. By constructing the depth-salinity mathematical model using the Gaussian process regression algorithm through explicit steps, this invention can accurately describe the nonlinear relationship between depth and salinity, laying a reliable data foundation for subsequent model correction and depth calculation.

[0080] S33 specifically includes:

[0081] S331: Extract each data record from the historical salinity data after normalization in S32, including depth. and normalized salinity value , where i represents the data sequence number, ensuring that the data has temporal consistency and spatial distribution consistency;

[0082] S332: Depth As the input variable of the model, the input vector X is constructed, and its expression is: Where n is the total number of data records, This represents the depth value of each record;

[0083] S333: Normalized salinity value As the output variable of the model, the output vector Y is constructed, and its expression is: ,in, This represents the salinity value corresponding to each record;

[0084] S334: Pair the input vector X and the output vector Y to form the training dataset T, which is expressed as follows: Ensure that each input depth data point is precisely matched with its corresponding salinity data.

[0085] S335: To improve the coverage of the training dataset, the training data is augmented; specifically, linear interpolation is used to fill in missing data within a continuous depth range. The interpolation calculation formula is as follows: ,in, and The depths are the front and rear points, respectively. and These are the corresponding salinity values. This is the interpolated salinity value. The depth of the interpolation point is used; the depth-salinity input-output pair is constructed through explicit steps, and the training dataset is enhanced by combining the interpolation algorithm, which ensures the integrity and continuity of the data and provides rich and accurate training samples for the Gaussian process regression algorithm.

[0086] The training depth-salinity mathematical model in S34 specifically includes:

[0087] S341: Initialize the Gaussian process regression model, selecting the radial basis function as the kernel function to measure the similarity between different depths in the training data;

[0088] S342: Based on the training dataset, calculate the similarity of each pair of depth data to generate a covariance matrix. Each element in the covariance matrix represents the correlation between two depth locations.

[0089] S343: Calculate the predicted salinity value at any test depth based on the covariance matrix and the salinity value in the training data, and obtain the predicted value corresponding to the test depth through the similarity between the training depth and the test depth and the observation noise parameter.

[0090] S344: The similarity parameter and observation noise parameter of the kernel function are optimized. The optimization process is based on maximizing the probability value of the training data under the model to ensure that the model can best fit the training data, thereby obtaining the depth-salinity mathematical model.

[0091] The specific training steps for training the depth-salinity mathematical model are as follows:

[0092] First, the Gaussian process regression model is initialized, and the radial basis function (RBF) is selected as the kernel function. The kernel function is used to measure the similarity between different depths, and the calculation formula is as follows: ,in, and This represents the i-th and j-th depth positions in the depth dataset. The length scale parameter of the kernel function is used to control the decay rate of similarity between depths;

[0093] Then, based on the training dataset T, the covariance matrix K of the Gaussian process regression is constructed, and the matrix elements are... Indicates depth and depth Similarities between them: , where i and j represent any two records in the dataset;

[0094] Next, the test depth is calculated using the covariance matrix K and the salinity values ​​S from the training data. Predicted salinity The formula is: ,in, Indicates test depth The kernel function value vector of the training depth set, The variance of the observation noise is represented by I, which represents the identity matrix.

[0095] Subsequently, the hyperparameters of the kernel function are optimized using maximum likelihood estimation. and observation noise variance This maximizes the likelihood function of the training dataset under a given model. The formula for calculating the likelihood function is: Where L represents the likelihood function. The determinant of the covariance matrix. The number of records in the training dataset;

[0096] Finally, based on the optimized hyperparameters and By using the depth-salinity pairs in the training dataset T, a Gaussian process regression model is trained, thus obtaining the depth-salinity mathematical model.

[0097] S4 specifically includes:

[0098] S41: By deploying water temperature sensors and ocean current velocity sensors, it collects real-time water temperature and ocean current velocity data of the target area; the water temperature sensor is used to accurately monitor changes in seawater temperature; the ocean current velocity sensor is used to measure the flow velocity at the ocean surface or different depths; at the same time, the collected data is preprocessed to remove noise and complete missing data to ensure data integrity.

[0099] S42: Based on preprocessed environmental data, analyze the effects of water temperature and ocean current velocity on salinity changes, and determine the relationship between environmental data and the depth-salinity model; where water temperature affects the solubility of salinity, and ocean current velocity affects the mixing and distribution of salinity;

[0100] S43: Based on the analysis results of S42, adjust the kernel function and related hyperparameters in the depth-salinity model; specifically, adjust the temperature factor in the kernel function according to water temperature changes to optimize the model's response under different water temperatures; and dynamically adjust the current velocity factor in the kernel function according to changes in ocean current velocity, so that the depth-salinity model can better adapt to the changes in salinity distribution caused by changes in ocean current velocity; by collecting environmental data in real time (such as water temperature and ocean current velocity) and adjusting the kernel function and hyperparameters in the model, the depth-salinity model can adapt to different marine environmental conditions, thereby improving the accuracy and flexibility of salinity prediction and ensuring that the prediction model can maintain high accuracy and reliability in various actual environments.

[0101] S42 specifically includes:

[0102] S421: Based on the preprocessed environmental data, extract time-series data of water temperature T and ocean current velocity v, corresponding to different depth locations. , forming data pairs ;

[0103] S422: Analyze the effect of water temperature T on salinity solubility, describing the relationship between water temperature changes and salinity through salinity solubility. The formula is: ,in, This indicates the effect of water temperature on salinity solubility. Indicates standard water temperature The salinity value at that time Temperature is the influencing factor, describing the rate at which water temperature reduces the salinity solubility.

[0104] S423: Analyze the influence of ocean current velocity v on salinity mixing and distribution. Using a salinity diffusion model from fluid dynamics, describe the effect of ocean current velocity changes on the salinity gradient. The expression is: ,in, This represents the rate of change in salinity distribution caused by ocean current velocity. Ocean current influencing factors This represents the depth gradient of salinity.

[0105] S424: Integrate the effects of water temperature and ocean current velocity into a comprehensive model, constructing a comprehensive expression for the relationship between environmental data and salinity changes: ,in, This represents the salinity value at a specified depth d under the combined effects of water temperature and ocean current velocity. Through detailed analysis of the influence of water temperature on salinity solubility and ocean current velocity on salinity distribution, an accurate relationship expression between environmental data and the depth-salinity model was constructed. This provides a clear theoretical basis and practical application scheme for dynamically correcting model parameters, ensuring that the model adapts to real-time changes in the marine environment.

[0106] S43 specifically includes:

[0107] S431: Based on the comprehensive relationship expression of salinity changes using environmental data, extract the influence of environmental factors on the salinity model, specifically including water temperature factors. and flow rate factor These are used to describe the degree of influence of water temperature and ocean current velocity on salinity;

[0108] S432: Define the kernel function for the Gaussian process regression model and temperature factor and flow rate factor The hyperparameter of the Gaussian process kernel function is expressed as:

[0109] The parameters of the kernel function are initialized based on environmental conditions.

[0110] S433: Based on the comprehensive relationship expression of salinity changes according to environmental data, dynamically adjust the hyperparameters in the kernel function; specifically, adjust the temperature factor through an optimization algorithm that maximizes the log-likelihood function. and flow rate factor To minimize the prediction error, the formula is: ,in, Let be the log-likelihood function, y be the observed salinity changes, and K be the kernel matrix. It is the sample size;

[0111] S434: The optimal temperature factor is obtained by maximizing the log-likelihood function. and flow rate factor The optimized parameters are then fed back into the Gaussian process regression model to dynamically correct the depth-salinity mathematical model. By dynamically adjusting the temperature and current factors in the Gaussian process regression model and optimizing the kernel function hyperparameters, the above steps can significantly improve the adaptability and prediction accuracy of the salinity-depth model to different environmental conditions, providing a scientific basis for depth-salinity prediction in complex marine environments.

[0112] S5 specifically includes:

[0113] S51: Input the preprocessed salinity data from S2 into the dynamically corrected depth-salinity mathematical model. This model is constructed and dynamically adjusted by the Gaussian process regression algorithm and can describe the nonlinear relationship between depth and salinity.

[0114] S52: Utilize the inverse calculation function of the depth-salinity mathematical model to calculate the corresponding seawater depth based on the salinity value;

[0115] S53: Perform confidence interval analysis on the calculation results. Quantify the uncertainty of the calculation results through the model's covariance matrix, and generate predicted values ​​and confidence intervals for each depth to ensure the reliability of the results.

[0116] S54: All predicted depth results are organized into a depth set and matched with actual salinity data to output the final salinity-depth relationship table; by inputting the preprocessed salinity data into the dynamically corrected depth-salinity mathematical model, combined with the model's inverse calculation and confidence interval analysis, the corresponding seawater depth information can be accurately derived, providing precise data support for marine environmental monitoring and salinity change analysis.

[0117] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0118] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for acquiring data at different depths of seawater based on a salinity sensor, characterized in that, Includes the following steps: S1: Deploy salinity sensors at different depths in the target ocean to collect real-time salinity data of seawater at different depths; S2: Preprocess the collected salinity data, including noise removal, outlier removal, and missing data interpolation; S3: Based on long-term historical salinity data, a depth-salinity mathematical model is constructed using the Gaussian process regression algorithm; S4: Real-time collection of environmental data, including water temperature and ocean current velocity, and dynamic correction of the parameters of the depth-salinity model based on the environmental data, adjusting the kernel function and related hyperparameters in the mathematical model of depth and salinity to adapt to the real-time changing marine environment; S4 specifically includes: S41: Real-time data on water temperature and ocean current velocity in the target area are collected through deployed water temperature sensors and ocean current velocity sensors; the water temperature sensors are used to accurately monitor changes in seawater temperature; the ocean current velocity sensors are used to measure the flow velocity at the ocean surface or at different depths; at the same time, the collected data is preprocessed to remove noise and complete missing data to ensure data integrity. S42: Based on preprocessed environmental data, analyze the effects of water temperature and ocean current velocity on salinity changes, and determine the relationship between environmental data and the depth-salinity model; where water temperature affects the solubility of salinity, and ocean current velocity affects the mixing and distribution of salinity; S43: Based on the analysis results of S42, adjust the kernel function and related hyperparameters in the depth-salinity model; specifically, adjust the temperature factor in the kernel function according to water temperature changes to optimize the model's response at different water temperatures; and dynamically adjust the velocity factor in the kernel function according to changes in ocean current velocity. S42 specifically includes: S421: Extract water temperature based on preprocessed environmental data. and ocean current speed Time series data; S422: Analyze water temperature The effect of salinity on solubility is described by the relationship between water temperature changes and salinity, using the formula: ,in, This indicates the effect of water temperature on salinity solubility. Indicates standard water temperature The salinity value at that time Temperature-related factors; S423: Analyzing Ocean Current Velocity The effect of salinity mixing and distribution is described using a salinity diffusion model from fluid dynamics, which illustrates the influence of ocean current velocity changes on the salinity gradient. The expression is as follows: ,in, This represents the rate of change in salinity distribution caused by ocean current velocity. Ocean current influencing factors This represents the depth gradient of salinity. S424: Integrate the effects of water temperature and ocean current velocity into a comprehensive model, constructing a comprehensive expression for the relationship between environmental data and salinity changes: ,in, Indicates at a specified depth The salinity value is determined by the combined effects of water temperature and ocean current velocity at different locations. S5: Input the preprocessed salinity data from S2 into the dynamically corrected depth-salinity mathematical model to calculate the corresponding seawater depth information.

2. The method for acquiring data at different depths of seawater based on a salinity sensor according to claim 1, characterized in that, S1 specifically includes: S11: Based on the water depth distribution of the target sea area, determine the sampling depth location, specifically selecting several depth points in the surface layer (0-50 meters), the middle layer (50-200 meters), and the deep layer (below 200 meters). S12: Deploy salinity sensors at each selected depth point, with each salinity sensor secured at the corresponding depth by a mooring cable or anchoring device; S13: Utilize the depth gauge built into the salinity sensor to monitor and record its depth information in real time, ensuring that the collected salinity data matches the corresponding depth one by one.

3. The method for acquiring data at different depths of seawater based on a salinity sensor according to claim 1, characterized in that, S2 specifically includes: S21: Noise removal processing is performed on the collected salinity data. A bandpass filter is used to filter the salinity signal to remove high-frequency noise and low-frequency interference. S22: Remove outliers from salinity data. Use the Z-Score algorithm to detect the salinity value of each data point. If the Z value of a data point exceeds the set threshold, the data point is considered an outlier and is removed. S23: Perform interpolation to complete missing data. Use linear interpolation or spline interpolation to fill in the missing salinity data in the time series to ensure data continuity and consistency.

4. The method for acquiring data at different depths of seawater based on a salinity sensor according to claim 1, characterized in that, S3 specifically includes: S31: Obtain long-term historical salinity data for the target sea area, specifically including salinity observations at different times and depths, covering a period of at least two years; S32: Normalize historical salinity data to bring salinity data at different depths into a standardized range. Specifically, use the minimum-maximum normalization method to map salinity values ​​to the interval [0, 1] to eliminate the magnitude differences between data at different depths and during observation periods. S33: Based on the historical salinity data after normalization in S32, use the depth of each data record as the input variable and salinity as the output variable to construct a depth-salinity input-output pair to form a training dataset. S34: Based on the training dataset formed in S33, the depth-salinity mathematical model is trained using the Gaussian process regression algorithm. Specifically, the radial basis function is selected as the kernel function, and the hyperparameters of the kernel function are optimized by the maximum likelihood estimation method, so that the depth-salinity mathematical model can capture the nonlinear relationship between depth and salinity.

5. The method for acquiring data at different depths of seawater based on a salinity sensor according to claim 4, characterized in that, Specifically, S33 includes: S331: Extract each data record from the historical salinity data after normalization in S32, including depth. and normalized salinity value ,in Indicates the sequence number of the data; S332: Depth As input variables of the model, construct the input vector. Its expression is: ,in, The total number of data records. This represents the depth value of each record; S333: Normalized salinity value As the output variable of the model, construct the output vector. Its expression is: ,in, This represents the salinity value corresponding to each record; S334: Input vector and output vector Pairing to form a training dataset Its expression is: ; S335: To improve the coverage of the training dataset, the training data is augmented; specifically, a linear interpolation method is used to fill in missing data within a continuous depth range.

6. The method for acquiring data at different depths of seawater based on a salinity sensor according to claim 5, characterized in that, S34 specifically includes: S341: Initialize the Gaussian process regression model, selecting the radial basis function as the kernel function to measure the similarity between different depths in the training data; S342: Based on the training dataset, calculate the similarity of each pair of depth data to generate a covariance matrix. Each element in the covariance matrix represents the correlation between two depth locations. S343: Calculate the predicted salinity value at any test depth based on the covariance matrix and the salinity value in the training data, and obtain the predicted value corresponding to the test depth through the similarity between the training depth and the test depth and the observation noise parameter. S344: The similarity parameter and observation noise parameter of the kernel function are optimized. The optimization process is based on maximizing the probability value of the training data under the model to ensure that the model can best fit the training data, thereby obtaining the depth-salinity mathematical model.

7. The method for acquiring data at different depths of seawater based on a salinity sensor according to claim 1, characterized in that, Specifically, S43 includes: S431: Based on the comprehensive relationship expression of salinity changes using environmental data, extract the influence of environmental factors on the salinity model, specifically including water temperature factors. and flow rate factor ; S432: Define the kernel function of the Gaussian process regression model and include the temperature factor. and flow rate factor As a hyperparameter of the Gaussian process kernel function; S433: Based on the comprehensive relationship expression of salinity changes according to environmental data, dynamically adjust the hyperparameters in the kernel function; specifically, adjust the temperature factor through an optimization algorithm that maximizes the log-likelihood function. and flow rate factor To minimize prediction error; S434: The optimal temperature factor is obtained by maximizing the log-likelihood function. and flow rate factor The optimized parameters are then fed back into the Gaussian process regression model to dynamically correct the depth-salinity mathematical model.

8. The method for acquiring data at different depths of seawater based on a salinity sensor according to claim 1, characterized in that, S5 specifically includes: S51: Input the preprocessed salinity data from S2 into the dynamically corrected depth-salinity mathematical model; S52: Utilize the inverse calculation function of the depth-salinity mathematical model to calculate the corresponding seawater depth based on the salinity value; S53: Perform confidence interval analysis on the calculation results. Quantify the uncertainty of the calculation results through the model's covariance matrix, and generate predicted values ​​and confidence intervals for each depth to ensure the reliability of the results. S54: Organize all predicted depth results into a depth set, match it with the actual salinity data, and output the final salinity-depth relationship table.

Images