A method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors.

Abstract

This invention relates to the field of environmental monitoring technology, specifically to a method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors. The method includes: S1: collecting nearshore environmental samples to obtain a dataset of physicochemical factors; S2: calculating the coupling strength between each physicochemical factor based on the dataset, generating a coupling strength matrix; S3: using principal component analysis to reduce the dimensionality of the coupling strength matrix and determine a list of key influencing factors; S4: analyzing the contribution of each dominant factor to the denitrification rate based on the list of key influencing factors, generating contribution analysis results; S5: outputting an influencing factor analysis report on the nearshore denitrification process based on the contribution analysis results, including a ranking of key influencing factors and optimization suggestions. This invention provides intelligent, data-driven auxiliary decision support for nearshore ecological restoration and water quality improvement, and has good prospects for widespread application.

Description

Technical Field

[0001] This invention relates to the field of environmental monitoring technology, and in particular to a method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors. Background Technology

[0002] In coastal waters, the enrichment of nitrogen nutrients has become one of the problems affecting water quality and ecosystem health. Denitrification, as an important denitrification pathway in the nitrogen cycle, can reduce nitrates in water to nitrogen gas, thereby removing nitrogen. In nearshore areas, due to strong hydrodynamic disturbances, complex pollution sources, and significant water stratification, the denitrification process is affected by the coupling of multiple physicochemical factors, including temperature, salinity, dissolved oxygen, and nitrate concentration. However, existing studies mostly focus on correlation analysis of single factors or empirical judgments, lacking systematic methods for mining factor coupling mechanisms, making it difficult to accurately identify the main controlling factors driving denitrification changes.

[0003] Furthermore, existing methods for quantitatively assessing influencing factors often employ univariate regression, subjective weighting, or empirical models, which suffer from strong subjectivity, unstable results, and insufficient explanatory power, making them difficult to adapt to the management needs under heterogeneous aquatic environmental conditions in different nearshore areas. There is also a lack of a systematic analytical process that starts from actual monitoring data and takes into account coupling analysis, dimensionality reduction, and regression discrimination, leading to a reliance on experience and poor generalization in the formulation of denitrification control strategies. Therefore, there is an urgent need to propose an analytical method oriented towards multi-factor coupling mechanisms, with the ability to screen factors and quantify contributions, to support the scientific and precise management of nitrogen in nearshore waters. Summary of the Invention

[0004] This invention provides a method for analyzing the influencing factors of nearshore denitrification processes by coupling physicochemical factors.

[0005] A method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors, comprising the following steps:

[0006] S1: Collect nearshore environmental samples and obtain a physicochemical factor dataset, which includes temperature data, salinity data, dissolved oxygen data, and nitrate concentration data;

[0007] S2: Based on the physicochemical factor dataset, calculate the coupling strength between each physicochemical factor and generate a coupling strength matrix;

[0008] S3: Based on the coupling strength matrix, the principal component analysis method is used to reduce the dimensionality of the coupling strength matrix and determine the list of key influencing factors of the denitrification process. The list of key influencing factors includes at least one dominant factor.

[0009] S4: Based on the list of key influencing factors, analyze the contribution of each dominant factor to the denitrification rate and generate contribution analysis results;

[0010] S5: Based on the contribution analysis results, output an analysis report on the influencing factors of the nearshore denitrification process. The analysis report includes the ranking of key influencing factors and optimization suggestions.

[0011] Optionally, S1 includes:

[0012] S11: Based on the hydrodynamic conditions and pollution source distribution of the target nearshore area, multiple sampling points are set up using the grid method or the cross-sectional method.

[0013] S12: Use a water quality sampler to collect water samples from the surface, middle and bottom water layers at each sampling point;

[0014] S13: Use a temperature, salinity, and depth meter to simultaneously measure and record the temperature and salinity data of each water layer on-site;

[0015] S14: The collected water samples were stored at low temperature and protected from light, and transported to the laboratory. The dissolved oxygen data were determined by Winkler titration, and the nitrate concentration data were determined by cadmium column reduction method.

[0016] S15: Verify and integrate the measurement data with the laboratory analysis data to construct a physicochemical factor dataset.

[0017] Optionally, S2 includes:

[0018] S21: Standardize the physicochemical factor dataset to eliminate the influence of dimensions;

[0019] S22: Use grey relational analysis to calculate the grey relational degree between any two physicochemical factors, which is used as the coupling strength.

[0020] S23: Fill the grey relational degree between all pairs of physicochemical factors into a symmetric matrix to generate the coupling strength matrix.

[0021] Optionally, S22 includes:

[0022] S221: For any two physicochemical factors, based on their standardized sample sequences, calculate the absolute value of the difference at different sample points, and construct the corresponding correlation coefficient sequence according to the grey relational analysis method.

[0023] S222: Average the correlation coefficient sequence over all sample points to obtain the grey correlation degree between the two physicochemical factor pairs, and use the grey correlation degree as the coupling strength between the two physicochemical factors.

[0024] Optionally, S3 includes:

[0025] S31: Perform principal component analysis on the coupling strength matrix of physicochemical factors, extract eigenvalues ​​and their corresponding eigenvectors, and identify the coupling structure characteristics reflected by each principal component;

[0026] S32: Select principal components with eigenvalues ​​greater than a preset threshold from the principal components, and determine the dominant principal components based on their corresponding variance contribution rates;

[0027] S33: Based on the dominant principal components, calculate the principal component loadings of each physicochemical factor and extract the absolute values ​​of the loadings of each factor;

[0028] S34: Identify physicochemical factors whose absolute loading values ​​are greater than the principal component loading threshold as dominant factors and include them in the list of key influencing factors.

[0029] Optionally, S32 includes:

[0030] S321: In the principal component analysis results, extract all principal components whose eigenvalues ​​are greater than a preset threshold, where the preset threshold is used to measure the information interpretability of the principal components.

[0031] S322: Calculate the variance contribution rate of the retained principal components, sort them from high to low contribution rate, and select the principal component with the largest contribution rate as the dominant principal component.

[0032] Optionally, S4 includes:

[0033] S41: Construct a multiple linear regression model with the dominant factors in the list of key influencing factors as independent variables and the denitrification rate as the dependent variable;

[0034] S42: Fit the multiple linear regression model to obtain the standardized regression coefficients corresponding to each dominant factor; take the absolute value of the standardized regression coefficient of each dominant factor and calculate its proportion to the sum of the absolute values ​​of the standardized regression coefficients of all dominant factors. The proportion is the contribution of the corresponding dominant factor to the denitrification rate.

[0035] S43: Summarize the contributions of all dominant factors and generate contribution analysis results.

[0036] Optionally, the multiple linear regression model includes:

[0037] The dominant factor data and corresponding denitrification rate data in the list of key influencing factors are standardized to eliminate the dimensional differences between variables.

[0038] The least squares method was used to construct a multiple linear regression equation with the standardized dominant factors as independent variables and the denitrification rate as the dependent variable, and the standardized regression coefficients of each dominant factor were obtained by fitting the equation.

[0039] Optionally, S5 includes:

[0040] S51: Based on the contribution analysis results, sort all dominant factors from high to low contribution to generate a ranking of key influencing factors;

[0041] S52: For at least one dominant factor with the highest contribution in the ranking of the key influencing factors, generate corresponding environmental management optimization suggestions, the optimization suggestions including at least one of regulating the water body stratification structure, implementing targeted pollution source emission reduction or artificially enhancing water body reoxygenation;

[0042] S53: Integrate the ranking of the key influencing factors with the environmental management optimization recommendations, and supplement with a visual chart of contribution distribution to automatically generate an influencing factor analysis report of the nearshore denitrification process.

[0043] Optionally, the impact factor analysis report of the nearshore denitrification process includes a sorted list table, a bar chart or pie chart visualization of the contribution of each dominant factor, and recommended management suggestions for high-contribution factors.

[0044] The beneficial effects of this invention are:

[0045] This invention is the first to introduce grey relational analysis into the modeling of the coupling relationship between physicochemical factors in nearshore denitrification processes, effectively overcoming the limitations of traditional methods in analyzing single factors in isolation. By constructing a coupling strength matrix, the dynamic correlation characteristics between key factors such as temperature, salinity, dissolved oxygen, and nitrate are revealed, providing a high-dimensional structural basis for subsequent identification of influencing mechanisms, thereby improving the accuracy of the analysis of the main controlling factors of the denitrification process.

[0046] This invention extracts the dominant principal components in the coupling matrix through principal component analysis, and further uses a standardized multiple linear regression model to quantitatively evaluate the contribution of each dominant factor to the denitrification rate. Compared with traditional sensitivity analysis or empirical weighting methods, it has stronger comparability, repeatability and stability, and effectively improves the scientificity and robustness of the screening and ranking of key influencing factors.

[0047] This invention ultimately forms an impact factor analysis report system that integrates factor ranking, contribution visualization, and automatic generation of management suggestions. Based on the characteristics of denitrification driving factors in specific regions, it can output targeted suggestions for water body regulation, pollution reduction, or reoxygenation measures, providing intelligent, data-driven auxiliary decision support for nearshore ecological restoration and water quality improvement, and has good prospects for promotion and application. Attached Figure Description

[0048] 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.

[0049] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0050] Figure 2 This is a schematic diagram of principal component extraction according to an embodiment of the present invention. Detailed Implementation

[0051] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art may employ 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.

[0052] like Figures 1-2 As shown, a method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors includes the following steps:

[0053] S1: Collect nearshore environmental samples and obtain a physicochemical factor dataset, which includes temperature data, salinity data, dissolved oxygen data, and nitrate concentration data;

[0054] S1 specifically includes:

[0055] S11: Based on the hydrodynamic conditions (tidal current velocity, wave height distribution) and pollution source distribution in the target nearshore area, sampling points are deployed using either the grid method or the cross-sectional method. The total number of sampling points is set to [number missing]. Each sampling point is marked as ;

[0056] in, This represents the total number of sampling points, ranging from 5 to 20. This value is determined by considering the complexity of nearshore hydrodynamics and the density of pollution source distribution to ensure spatial representativeness and computational controllability. For the first Each sampling point is numbered, and the number is used to index the sampling points to meet the needs of subsequent matrix matching and spatial statistical analysis.

[0057] S12: Use a water quality sampler to sample the surface layer at each sampling point. Middle layer and the bottom layer Water samples are collected, and the collection of water samples from each sampling point is represented as follows:

[0058] ;

[0059] in, For water sample collection, Representing points respectively Surface, middle and bottom water samples.

[0060] S13: Use a temperature, salinity, and depth (CTD) meter to measure physical factors, including temperature, in each water layer sample at each sampling point. and salinity The CTD output is represented as follows:

[0061] ;

[0062] in, This indicates the water layer type. The denitrification process is closely related to vertical water quality differences. The three-layer design can capture the gradient across the entire water depth. For the first The sampling point The water temperature ranges from 5-35℃. The activity of denitrifying bacteria is highly temperature-dependent. The marine environment varies greatly throughout the four seasons, requiring full coverage. For the first The sampling point Salinity, ranging from 5 to 35 PSU, affects dissolved oxygen saturation and the structure of denitrifying bacteria communities. Extensive coverage of estuaries and open seas is required. Representing points respectively The surface, middle and bottom water temperatures, Representing points respectively The salinity of the surface, middle and bottom layers.

[0063] S14: Water samples from each layer were stored at low temperature and protected from light, and transported to the laboratory within 24 hours. Dissolved oxygen concentration was determined using the Winkler titration method. The results were as follows: (Unit: mg / L) The nitrate concentration was determined using the cadmium column reduction method, and the results were as follows: (Unit: mg / L), the output laboratory data is expressed as follows:

[0064] ;

[0065] in, For the first The sampling point Dissolved oxygen concentration in the layer For the first The sampling point The nitrate concentration should be between 0-12 mg / L. Dissolved oxygen is an inhibitor of denitrification; 0-2 mg / L is the favorable range for anaerobic conditions. The range should cover the transition from aerobic to anoxic environments. Representing points respectively Dissolved oxygen concentrations in the surface, middle, and bottom layers, Representing points respectively The nitrate concentrations in the surface, middle, and bottom layers, This is a collection of laboratory analysis results.

[0066] The cadmium column reduction method is a classic colorimetric analysis method commonly used to determine the nitrate content in water. Its core principle is to use copper-activated cadmium metal to selectively reduce nitrate in the water sample to nitrite, and then generate a pink compound with characteristic absorption in the visible light region through an azo reaction with a colorimetric reagent, thereby achieving quantitative analysis.

[0067] S15: Transfer in-situ measurement data Laboratory analysis data Corresponding integration forms the final physicochemical factor dataset. , is represented as:

[0068] ;

[0069] in, This is the total set of physicochemical factor datasets.

[0070] S2: Based on the physicochemical factor dataset, calculate the coupling strength between each physicochemical factor and generate a coupling strength matrix;

[0071] S2 specifically includes:

[0072] S21: To eliminate the dimensional influence between various physicochemical factors, range standardization is performed on each class of physicochemical factors in the original dataset, as follows:

[0073] ;

[0074] in, It is the first The physicochemical factor in the first The original values ​​at each sample, These are the standardized values, ranging from 0 to 1. , It is the first The minimum and maximum values ​​of each factor, and the standardized data matrix is ​​denoted as . , This refers to the total number of samples, i.e., the number of samples participating in standardization, with a value ranging from 9 to 12, to ensure analytical stability. The number of physicochemical factors is 4. The physicochemical factors include temperature, salinity, dissolved oxygen, and nitrate. After standardization, all factor values ​​are distributed in the [0,1] interval, which is beneficial to enhance the comparability of coupling strength and matrix symmetry.

[0075] S22: The grey relational analysis method is used to calculate the coupling strength between any two physicochemical factor sequences, defined as the grey relational degree. Let the first... The factor and the first The standardized sequences of the factors are as follows: and The corresponding grey relational coefficient and overall grey relational degree are expressed as follows:

[0076] ;

[0077] ;

[0078] in, It is the first The nth sample point The standardized values ​​of the physicochemical factors range from 0 to 1. Range standardization eliminates dimensional differences between factors, facilitating coupled comparisons. It is the first The first sample point and The correlation coefficient between the factors For the first and The grey relational degree of each factor, i.e., the coupling strength. The resolution coefficient is set to 0.5. The smaller the value, the more the overall correlation coefficient tends to average out. The larger the value, the more the differences are amplified; 0.5 is the empirically optimal value, balancing discriminability and stability. Indicates the number of sample points;

[0079] This step uses grey relational analysis to quantify the coupling strength between physicochemical factors, aiming to discover the coupling relationships of key factors affecting nearshore denitrification processes. It includes two aspects: standardized preprocessing and grey relational coefficient calculation.

[0080] First, because different physicochemical factors have different dimensions (e.g., temperature is in °C, salinity is in PSU, and dissolved oxygen and nitrate concentrations are in mg / L), direct comparison would lead to calculation errors due to differences in numerical scale. Therefore, a range standardization method is used to unify all factors to a uniform scale. Within the specified interval, the influence of dimensions is eliminated, making subsequent analyses comparable and objective.

[0081] Secondly, the grey relational coefficient from grey system theory is used to measure the similarity of the changing trends between two factor sequences. This method does not rely on the distribution form of variables or linear assumptions, making it suitable for systems with small samples, high uncertainty, and incomplete information, which perfectly matches the data characteristics of the complex coastal environment. Furthermore, by introducing a resolution coefficient... This is used to adjust the sensitivity to ensure that it is both discerning when dealing with small differences between factors, without over-amplifying noise.

[0082] Finally, the average grey relational coefficients of each pair of factors are summed to construct a symmetric coupling strength matrix, which is used for subsequent identification and analysis of key factors. This method has strong adaptability and interpretability, and also provides a structured input basis for subsequent visualization and multi-factor model analysis.

[0083] S23: All pairwise grey correlation between physicochemical factors Enter one Symmetric matrix This is the final generated coupling strength matrix, expressed as:

[0084] ;

[0085] in, for The coupling strength matrix, , ;

[0086] In a multi-factor system, the matrix form can comprehensively, intuitively, and clearly express the bilateral coupling relationship between various factors, which facilitates further analysis such as identification of dominant factors, cluster analysis, or visualization of coupling networks. This matrix not only supports quantitative ranking and screening, but also provides a data foundation and logical support for the system modeling and optimization of influencing factors, which meets the actual needs of the complex interaction of multiple factors in coastal denitrification research.

[0087] S3: Based on the coupling strength matrix, the principal component analysis method is used to reduce the dimension of the coupling strength matrix and determine the list of key influencing factors of the denitrification process. The list of key influencing factors includes at least one dominant factor.

[0088] S3 specifically includes:

[0089] S31: To be constructed Coupling strength matrix As input to principal component analysis, linear algebraic methods are used to extract the eigenvalues ​​and eigenvectors of the matrix, obtaining the coupling dimension represented by each principal component and its variance interpretability. Solve for the following feature structure, represented as:

[0090] ;

[0091] in, It is the first The eigenvalues ​​corresponding to each principal component range from 0.1 to 3.0. Is with The corresponding unit eigenvector represents the first... The direction of each principal component Principal component index number;

[0092] This step uses Principal Component Analysis (PCA) to reduce the dimensionality of the coupling strength matrix constructed in the previous stage, extracting the core principal components that reflect the coupling structure between physicochemical factors. The coupling strength matrix is ​​a comprehensive quantitative description of the correlation between factors, but in actual analysis, due to a certain degree of information redundancy and collinearity among factors, directly using the complete matrix will lead to a complex and difficult-to-interpret model. Therefore, PCA is used to perform eigenvalue decomposition on the matrix, projecting the original high-dimensional data onto a set of linearly uncorrelated principal component axes, thereby simplifying the variable structure and extracting the most representative coupling patterns.

[0093] This process calculates the eigenvalues ​​and eigenvectors of the matrix, assigning each principal component a specific information content. The information content is determined by the magnitude of the eigenvalue, while the eigenvector describes the projection intensity of that principal component onto each of the original factor directions. This approach not only effectively preserves the main coupling relationships but also eliminates noise interference and weakly correlated factors, facilitating the subsequent identification and ranking of dominant factors.

[0094] Principal component analysis (PCA) is a commonly used data dimensionality reduction technique that aims to extract a few new, uncorrelated variables (i.e., principal components) from multiple interrelated variables. These principal components retain as much information as possible from the original variables. In this step, PCA is used to extract the dominant coupling factor combinations controlling the denitrification process from the coupling strength matrix, thereby simplifying the variable system and highlighting key influencing factors. The mathematical essence of PCA is eigenvalue decomposition of the input matrix, a classic linear algebra operation used to find the invariant scaling properties of a square matrix in certain directions. By decomposing the matrix into multiple eigenvalue-eigenvector pairs, the main directions of change in the data can be identified, i.e., the factor combinations with the most prominent coupling characteristics.

[0095] S32: To avoid introducing principal components with limited information, an eigenvalue selection criterion is set, retaining only those that meet the criteria. The principal components are used for subsequent analysis, and the selected principal component set is represented as follows:

[0096] ;

[0097] To further clarify the importance of each principal component, the corresponding variance contribution rate is calculated and expressed as:

[0098] ;

[0099] in, For the first The variance contribution rate of each principal component represents the proportion of that principal component in the total information. The larger the value, the more information the principal component interprets. The largest principal component is taken as the dominant principal component, denoted as . , The total variance is the sum of the eigenvalues ​​of all principal components. Assuming the input matrix is ​​standardized, the total variance of principal component analysis is close to the sum of the original number of variables. To and The corresponding unit eigenvector, The set of principal components to be retained excludes weakly explanatory principal components and retains those with high information content, thereby improving analytical efficiency and robustness. To ensure that the retained principal components have at least the same explanatory power as one of the original factors, only those that meet the following criteria are retained. Principal components can both avoid information loss and eliminate redundant variables.

[0100] This scheme selects the dominant principal component with the strongest explanatory power for the overall information from multiple principal components, providing a basis for subsequent identification of key influencing factors. First, an eigenvalue threshold standard is set, that is, only principal components with eigenvalues ​​greater than 1 are retained. According to the Kaiser criterion in principal component analysis, it is believed that the principal components with eigenvalues ​​less than 1 contain less information than the original variables themselves and do not have the necessity to be retained independently, so they should be discarded.

[0101] To further measure the importance of each principal component among the retained principal components, a variance contribution rate index is introduced, which is the proportion of each principal component's eigenvalue to the sum of all eigenvalues. This index is used to characterize the proportion of information carried by the principal component. The higher the contribution rate, the more representative the coupling structure between the original variables is. Finally, the principal component with the largest contribution rate is selected as the dominant principal component, which best reflects the main coupling trend in the physicochemical factor system, laying the foundation for subsequent loading analysis and dominant factor identification.

[0102] S33: To evaluate the weight of each physicochemical factor in the dominant principal component, the absolute value of each component in the dominant principal component vector is extracted as the principal component loading of the factor, expressed as:

[0103] ;

[0104] in, Indicates the first The projection intensity of each physicochemical factor in the dominant principal component, the larger the value, the greater its contribution;

[0105] S34: To ultimately screen out the dominant factors that significantly affect the denitrification process, a principal component loading threshold is set. When the loading of a certain factor satisfies When a factor is considered the dominant factor, it is included in the list of key influencing factors, thus forming the final list of key influencing factors. These are the key influencing factors identified based on coupling relationships and information contribution, which can be used for explaining the denitrification process mechanism and optimizing control parameters, and are expressed as:

[0106] ;

[0107] in, Principal component loading thresholds for the dominant factor set (list of key influencing factors). The value range is 0.6-0.8. If it is set too low, many weak contributing factors will be included, causing the identification results to be generalized. If it is set too high, less important factors may be missed, affecting the completeness.

[0108] This step identifies the key factors that play a dominant role in the denitrification coupling mechanism based on the loading magnitude of each physicochemical factor in the principal component. The loading values ​​of each factor in the principal component vector are extracted, and their absolute values ​​are compared with a preset threshold. If the absolute value of the loading of a factor is greater than or equal to the threshold, it is determined to be the dominant factor.

[0109] Each component in the principal component vector represents the projection intensity of the original factor along that principal component direction, i.e., the factor's contribution or explanatory power. By setting a reasonable threshold, the minority variables with the most significant impact on the dominant structure can be effectively screened out, thus highlighting key factors and avoiding interference. This method has clear quantitative standards, is easy to operate and compare, and also conforms to the objective law that the dominant minority factors control the main processes in actual ecological or environmental systems.

[0110] S4: Based on the list of key influencing factors, analyze the contribution of each dominant factor to the denitrification rate and generate contribution analysis results;

[0111] S4 specifically includes:

[0112] S41: Using the dominant factor in the list of key influencing factors as the independent variable and the actual observed or simulated denitrification rate as the dependent variable, construct a linear model, expressed as:

[0113] ;

[0114] in, The denitrification rate (dependent variable) characterizes nitrogen conversion capacity and is the response target of the denitrification process. For the first One dominant factor (independent variable). This refers to the number of dominant factors included in the list of key influencing factors after principal component analysis and loading screening. The value ranges from 2 to 5, and is kept within a relatively small dimension to improve model stability. The intercept term of the regression model represents the baseline denitrification rate when all dominant factors are zero, and the value represents the strength and direction of the linear influence of the factor on the denitrification rate. Since the standardized values ​​are comparable, For the first The standardized regression coefficients of the dominant factors range from -2 to -2. These are residual terms with a mean of 0.

[0115] This step involves constructing a multiple linear regression model to quantitatively analyze the influence of each dominant factor among the key influencing factors on the denitrification rate. The dominant factors are used as independent variables in the model, and the actual observed or simulated denitrification rate is used as the dependent variable. A linear regression equation is then established to fit the explanatory power of each factor on the denitrification rate.

[0116] Before model construction, all variables were standardized to eliminate interference from differences in units, dimensions, and numerical scales in the comparison of regression coefficients, ensuring the comparability of each regression coefficient. The standardized regression coefficients directly reflect the linear influence strength of each factor. The model includes intercept and residual terms to describe the system's baseline response level and unexplained bias.

[0117] S42: The fitted regression coefficients Calculate its absolute value and then determine the normalized contribution ratio of this factor among all dominant factors, defining it as the contribution degree. , is represented as:

[0118] ;

[0119] in, For the first The contribution of each dominant factor to the denitrification rate is calculated, with the denominator being the sum of the absolute values ​​of all standardized regression coefficients, ensuring that the sum of the contributions equals 1. This indicates the relative strength of the factor's influence, ignoring the sign and focusing only on the magnitude;

[0120] This step, after establishing the multiple linear regression model, further quantifies the relative influence of each dominant factor on the denitrification rate based on the regression results. The absolute value of the standardized regression coefficient of each dominant factor is used as a measure of its influence strength, and this value is normalized to the sum of the absolute values ​​of the regression coefficients of all dominant factors, thus obtaining the relative contribution of each factor. The standardized regression coefficients have eliminated the influence of the variables' dimensions and numerical ranges, so their absolute values ​​can be directly used to measure the strength of the linear influence of each variable on the dependent variable. Simultaneously, the normalization process ensures that the sum of the contributions of all factors is 1, facilitating horizontal comparison and ranking among different factors.

[0121] S43: All The results are summarized and sorted from largest to smallest contribution to form a contribution analysis result set, represented as follows:

[0122] ;

[0123] in, This represents the set of contribution analysis results. For the first One dominant factor For the first The contribution of each dominant factor to the denitrification rate.

[0124] S5: Based on the contribution analysis results, output the impact factor analysis report of the nearshore denitrification process. The analysis report includes the ranking of key impact factors and optimization suggestions.

[0125] S5 specifically includes:

[0126] S51: Based on the contribution of the dominant factors, sort all the dominant factors from highest to lowest contribution to generate a ranking list of key influencing factors:

[0127] , ;

[0128] in, Indicates the first The dominant factor in positional ordering Its corresponding contribution;

[0129] S52: For at least one dominant factor with the largest contribution in the key impact factor ranking list, call the preset optimization strategy library to generate corresponding environmental management optimization suggestions. The optimization suggestions include at least one of the following measures:

[0130] Regulating the water stratification structure: By intervening in hydrodynamics or guiding mixing, weakening or strengthening thermosalinous stratification and improving physical exchange between upper and lower layers;

[0131] Implement pollution source reduction: Identify and control major nitrate input fluxes in the region (such as surface runoff and point source emissions).

[0132] Artificial enhancement of water reoxygenation: By increasing the dissolved oxygen concentration at the bottom through aeration, underwater aerators and other methods, the anaerobic environment is inhibited, and the denitrification pathway is indirectly regulated.

[0133] S53: Integrate the ranking results of key impact factors, the contribution values ​​of dominant factors, and corresponding optimization suggestions to construct a complete impact factor analysis report, including:

[0134] Sort list table;

[0135] Visualize the contribution of each dominant factor using a bar chart or pie chart;

[0136] Recommendations for management of high-contribution factors.

[0137] 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.

[0138] 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 analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors, characterized in that, Includes the following steps: S1: Collect nearshore environmental samples and obtain a physicochemical factor dataset, which includes temperature data, salinity data, dissolved oxygen data, and nitrate concentration data; S2: Based on the physicochemical factor dataset, calculate the coupling strength between each physicochemical factor and generate a coupling strength matrix; specifically including: S21: Standardize the physicochemical factor dataset to eliminate the influence of dimensions; S22: Use grey relational analysis to calculate the grey relational degree between any two physicochemical factors, which is used as the coupling strength. S23: Fill the gray correlation between all pairs of physicochemical factors into a symmetric matrix to generate the coupling strength matrix; S3: Based on the coupling strength matrix, principal component analysis is used to reduce the dimensionality of the coupling strength matrix and determine the list of key influencing factors for the denitrification process. This list of key influencing factors includes at least one dominant factor; specifically, it includes: S31: Perform principal component analysis on the coupling strength matrix of physicochemical factors, extract eigenvalues ​​and their corresponding eigenvectors, and identify the coupling structure characteristics reflected by each principal component; S32: Select principal components with eigenvalues ​​greater than a preset threshold from the principal components, and determine the dominant principal components based on their corresponding variance contribution rates; S33: Based on the dominant principal components, calculate the principal component loadings of each physicochemical factor and extract the absolute values ​​of the loadings of each factor; S34: Identify physicochemical factors whose absolute loading values ​​are greater than the principal component loading threshold as dominant factors and include them in the list of key influencing factors. S4: Based on the list of key influencing factors, analyze the contribution of each dominant factor to the denitrification rate and generate contribution analysis results; S5: Based on the contribution analysis results, output an analysis report on the influencing factors of the nearshore denitrification process. The analysis report includes the ranking of key influencing factors and optimization suggestions.

2. The method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors according to claim 1, characterized in that, S1 includes: S11: Based on the hydrodynamic conditions and pollution source distribution of the target nearshore area, multiple sampling points are set up using the grid method or the cross-sectional method. S12: Use a water quality sampler to collect water samples from the surface, middle and bottom water layers at each sampling point; S13: Use a temperature, salinity, and depth meter to simultaneously measure and record the temperature and salinity data of each water layer on-site; S14: The collected water samples were stored at low temperature and protected from light, and transported to the laboratory. The dissolved oxygen data were determined by Winkler titration, and the nitrate concentration data were determined by cadmium column reduction method. S15: Verify and integrate the measurement data with the laboratory analysis data to construct a physicochemical factor dataset.

3. The method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors according to claim 1, characterized in that, S22 includes: S221: For any two physicochemical factors, based on their standardized sample sequences, calculate the absolute value of the difference at different sample points, and construct the corresponding correlation coefficient sequence according to the grey relational analysis method. S222: Average the correlation coefficient sequence over all sample points to obtain the grey correlation degree between the two physicochemical factor pairs, and use the grey correlation degree as the coupling strength between the two physicochemical factors.

4. The method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors according to claim 1, characterized in that, S32 includes: S321: In the principal component analysis results, extract all principal components whose eigenvalues ​​are greater than a preset threshold, where the preset threshold is used to measure the information interpretability of the principal components. S322: Calculate the variance contribution rate of the retained principal components, sort them from high to low contribution rate, and select the principal component with the largest contribution rate as the dominant principal component.

5. The method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors according to claim 1, characterized in that, S4 includes: S41: Construct a multiple linear regression model with the dominant factors in the list of key influencing factors as independent variables and the denitrification rate as the dependent variable; S42: Fit the multiple linear regression model to obtain the standardized regression coefficients corresponding to each dominant factor; take the absolute value of the standardized regression coefficient of each dominant factor and calculate its proportion to the sum of the absolute values ​​of the standardized regression coefficients of all dominant factors. The proportion is the contribution of the corresponding dominant factor to the denitrification rate. S43: Summarize the contributions of all dominant factors and generate contribution analysis results.

6. The method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors according to claim 5, characterized in that, The multiple linear regression model includes: The dominant factor data and corresponding denitrification rate data in the list of key influencing factors are standardized to eliminate the dimensional differences between variables. The least squares method was used to construct a multiple linear regression equation with the standardized dominant factors as independent variables and the denitrification rate as the dependent variable, and the standardized regression coefficients of each dominant factor were obtained by fitting the equation.

7. The method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors according to claim 5, characterized in that, S5 includes: S51: Based on the contribution analysis results, sort all dominant factors from high to low contribution to generate a ranking of key influencing factors; S52: For at least one dominant factor with the highest contribution in the ranking of the key influencing factors, generate corresponding environmental management optimization suggestions, the optimization suggestions including at least one of regulating the water body stratification structure, implementing targeted pollution source emission reduction or artificially enhancing water body reoxygenation; S53: Integrate the ranking of the key influencing factors with the environmental management optimization recommendations, and supplement with a visual chart of contribution distribution to automatically generate an influencing factor analysis report of the nearshore denitrification process.

8. The method for analyzing the influencing factors of nearshore denitrification processes coupled with physicochemical factors according to claim 7, characterized in that, The impact factor analysis report of the nearshore denitrification process includes a sorted list table, a bar chart or pie chart visualization of the contribution of each dominant factor, and recommended management suggestions for high-contribution factors.

Images