Ecological damage evaluation method for deep-sea mining disturbance based on data analysis

Abstract

The invention relates to the technical field of ecological environment monitoring and evaluation, in particular to a deep-sea mining disturbance ecological damage evaluation method based on data analysis, which comprises the following steps of: S1, simultaneously acquiring data of various disturbance factors in a direct influence area and an indirect influence area of mining disturbance; s2, extracting characteristic parameters of each disturbance factor to obtain a characteristic set of each factor; s3, identifying factor combinations with interaction; s4, constructing a multi-factor fused ecological damage assessment model, and generating an ecological damage value; s5, dividing ecological damage grades; when the damage value reaches or exceeds the threshold value of the corresponding grade, a corresponding early warning mechanism is triggered, and a corresponding ecological protection suggestion is generated. According to the method, through multi-factor fusion analysis and a grading early warning mechanism, accurate evaluation and timely response to deep-sea mining disturbance ecological damage are realized, and the effectiveness of ecological protection measures is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological environment monitoring and evaluation, and in particular to an ecological damage evaluation method for deep-sea mining disturbance based on data analysis. Background Art

[0002] With the development of deep-sea mining technology, the demand for the exploitation of deep-sea mineral resources such as polymetallic nodules is increasing. However, deep-sea mining disturbances can cause potential damage to the marine ecosystem. In particular, the operation of mining equipment and excavation activities can disturb seabed sediments, release chemical components, and destroy the habitats of benthic organisms. These disturbance factors involve complex physical, chemical, and biological effects, which may cause structural damage and functional degradation of the ecosystem.

[0003] Existing ecological damage assessment technologies often have the following problems: First, there is a lack of comprehensive analysis of multiple disturbance factors, making it difficult to quantify the interactions among physical, chemical and biological disturbance factors and their overall impact on the ecosystem; second, there is a lack of a systematic ecological damage classification and early warning mechanism, making it impossible to accurately and promptly take corresponding protective measures at different stages of the damage; third, in terms of the generation of ecological damage values, there is a lack of models based on multi-factor fusion, making it difficult to accurately reflect the multi-dimensional ecological impacts of mining disturbances. Summary of the invention

[0004] Based on the above objectives, the present invention provides an ecological damage assessment method for deep-sea mining disturbance based on data analysis.

[0005] A method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis includes the following steps: S1: In the direct and indirect impact areas of mining disturbance, data on multiple disturbance factors are collected simultaneously, including physical data, chemical data, and biological data of water bodies and sediments to form an initial data set of multi-factor disturbance; S2: Preprocess the data of various disturbance factors collected by S1, extract the characteristic parameters of each disturbance factor by principal component analysis and wavelet decomposition, and obtain the characteristic set of each factor; S3: Based on the feature set extracted in S2, the association analysis algorithm is used to analyze the characteristic parameters of each disturbance factor, calculate the correlation between physical, chemical and biological factors, and identify the combination of factors with interactions; S4: Based on the correlation analysis results of S3, a multi-factor integrated ecological damage assessment model is constructed to comprehensively assess the overall impact of various factor combinations on the ecosystem and generate ecological damage values; S5: Based on the ecological damage value generated by S4, set the damage threshold of the ecosystem to different combinations of disturbance factors and divide the ecological damage level; when the damage value reaches or exceeds the threshold of the corresponding level, trigger the corresponding early warning mechanism and generate corresponding ecological protection recommendations.

[0006] Optionally, the S1 specifically includes: S11: According to the geographical scope and equipment parameters of mining activities, determine a number of fixed sampling points in the direct impact area and the indirect impact area, and define the same sampling range at each sampling point to ensure the coverage and consistency of data collection; S12: At each sampling point in the direct impact zone and the indirect impact zone, use a multi-parameter water quality sensor to collect physical data of the water body, including temperature and turbidity, and use a sediment sampler to obtain physical data of the sediment to comprehensively record the physical disturbance characteristics; S13: Use dissolved oxygen monitors, conductivity sensors, and metal ion concentration detection instruments to obtain water and sediment chemical data at each sampling point, including dissolved oxygen concentration and target metal ion concentration; S14: Collect benthic samples through benthic trawl sampling equipment, measure the benthic species and community density at each sampling point, and obtain biological data; S15: The collected physical, chemical and biological data are classified and summarized according to the sampling location and time sequence to generate a multi-factor disturbance initial data set to provide basic data for subsequent analysis.

[0007] Optionally, the S2 specifically includes: S21: Standardize the collected physical, chemical and biological data, and use data normalization or standardization algorithms to adjust each data item to the same order of magnitude to eliminate the dimensional differences between different disturbance factors; S22: Perform principal component analysis on the standardized physical, chemical and biological data, calculate the covariance matrix of each data item, and obtain the principal components through eigenvalue decomposition; and select the first few principal components whose cumulative contribution rate reaches the preset threshold, and compress the original data into low-dimensional feature data containing the main feature information to reduce the data dimension; S23: performing wavelet decomposition on the low-dimensional feature data obtained in S22, and using the selected Daubechies wavelet to decompose the time series data into components of different scales; obtaining the variation characteristics of the data in different frequency bands through multi-scale analysis, and extracting characteristic parameters reflecting the disturbance intensity and frequency changes; S24: Screen the characteristic parameters extracted by principal component analysis and wavelet decomposition, retain the characteristic parameters of the disturbance factors that have the greatest impact on ecosystem damage, and classify them into characteristic sets of physical, chemical and biological factors.

[0008] Optionally, the S23 specifically includes: S231: select Daubechies wavelet as wavelet basis function, and set the scale layer number of wavelet decomposition to decompose the collected time series data; S232: Decomposing the time series data into multiple components of different scales; the decomposed data includes high-frequency detail components and low-frequency approximate components, wherein the high-frequency detail components are used to capture the instantaneous change characteristics of the disturbance, and the low-frequency approximate components are used to reflect the overall change trend; S233: Calculate the scale factor of each detail component to quantify the intensity of the disturbance at each scale; S234: Based on the scale factor of each detail component, by setting a threshold of the disturbance intensity, a component with significant disturbance intensity is extracted as a feature parameter.

[0009] Optionally, the S24 specifically includes: S241: For each characteristic parameter extracted by principal component analysis and wavelet decomposition, calculate its influence on ecosystem damage. The influence is measured by the variance contribution rate of each characteristic parameter. The calculation formula is: ,in, represents the influence of the i-th feature parameter, represents the variance of the i-th feature parameter, is the total variance of all feature parameters; S242: based on the influence calculated in S241, all feature parameters are sorted by influence, and feature parameters whose influence exceeds a preset threshold are retained; S243: Classify and group the screened characteristic parameters according to physical, chemical and biological factors.

[0010] Optionally, the S3 specifically includes: S31: Pair the physical, chemical and biological characteristic parameters in pairs and calculate the Pearson correlation coefficient between each pair of characteristic parameters to quantify the linear correlation between the two. The calculation formula of the Pearson correlation coefficient is: ,in, is the correlation coefficient between the characteristic parameters x and y, where x and y represent the values ​​of the two characteristic parameters respectively. and is the average value of the corresponding parameter; The value range is from -1 to 1. The closer the value is to 1 or -1, the stronger the linear correlation between the two parameters. S32: Perform a significance test on the calculated Pearson correlation coefficient to determine whether the correlation reaches a significant level; the significance test is determined by calculating the t value and comparing it with the set critical value level. The calculation formula of the t value is: , where t is the statistical value of the significance test, is the Pearson correlation coefficient between the two characteristic parameters, n is the number of samples, and the calculated t value is compared with the critical t value. If the absolute value of the t value is greater than the critical value, it is considered that the correlation reaches a significant level; S33: All pairs of characteristic parameters with significant correlation are used to form a correlation matrix, and the Pearson correlation coefficient value of each pair of characteristic parameters is used as an element in the correlation matrix to form a matrix, the expression is: , where R is the incidence matrix, is the significant correlation coefficient between characteristic parameter i and characteristic parameter j; when When , the diagonal elements are 1, indicating the complete correlation of the parameters themselves; S34: Based on the correlation matrix, characteristic parameter combinations with high correlation coefficient values ​​are screened, and combinations with significant correlation among physical, chemical and biological characteristic parameters are identified as combinations of factors with interaction.

[0011] Optionally, the S34 specifically includes: S341: To select the characteristic parameter combination with high correlation, set the threshold of correlation coefficient , S342: Traverse each element in the correlation matrix R and select all elements that meet the conditions The characteristic parameter combination of represents the correlation coefficient between characteristic parameter i and characteristic parameter j; when Greater than or equal to threshold When , the characteristic parameter pair is recorded and marked as a high correlation combination; S343: Classify the screened high-correlation characteristic parameter combinations according to the three factors of physics, chemistry and biology, assign each characteristic parameter to its corresponding factor category, and form a cross-category combination table.

[0012] Optionally, the S4 specifically includes: S41: Based on the combination of highly correlated characteristic parameters identified in S3, determine the main input variables of physical, chemical and biological factors and construct the input variable set of the model; S42: Use the analytic hierarchy process to assign weights to each input variable and calculate the relative importance of each pair of factors in the ecological damage assessment by expert scoring or empirical weight setting; S43: Based on the determined input variables and weights, a multi-factor fusion ecological damage assessment model is constructed, and the ecological damage value of each factor combination to the ecosystem is calculated.

[0013] Optionally, the S43 specifically includes: S431: Taking the determined physical, chemical and biological input variables as the input layer of the model, a multi-factor fusion ecological damage assessment model is constructed, and the expression is: , where D is the comprehensive ecological damage value of the ecosystem, is the weight of the ith factor, is the characteristic parameter value of the i-th factor, is the total number of factors; S432: multiplying the characteristic parameter value of each input variable by its corresponding weight to calculate the weighted damage contribution value of each factor; S433: To ensure the comparability of ecological damage values, the calculated damage values ​​are normalized to the range .

[0014] Optionally, the S5 specifically includes: S51: Set the threshold of ecological damage classification based on historical ecological monitoring data and expert experience, and define corresponding thresholds for each damage level of the ecosystem; the damage threshold is set by using a hierarchical setting method, dividing the ecological damage value D into four levels, namely slight L1, moderate L2, severe L3 and extremely severe L4; the corresponding damage threshold is , , , and meet the following conditions: ; ; ; ;in, , , is the damage level threshold of the ecosystem; S52: Compare the ecological damage value generated by S4 with the classification threshold set in S51 to determine the current damage level of the ecosystem; according to the range of the damage value, divide it into corresponding damage levels L1, L2, L3 or L4 to determine the current ecological risk level; S54: When the calculated ecological damage value reaches or exceeds the set damage level threshold, the corresponding early warning mechanism is triggered and ecological protection recommendations are generated; the early warning mechanism will determine the early warning level according to the current ecological damage level, and the early warning level corresponds to the ecological damage level, L1 to L4 correspond to low, medium, high and extremely high warnings respectively; the protection recommendations include measures to slow down, suspend and repair the ecological environment of mining activities.

[0015] Beneficial effects of the present invention: The present invention, through multi-factor fusion analysis technology, can systematically integrate the physical, chemical and biological factors generated in the mining disturbance process, and construct an ecological damage assessment model to conduct a comprehensive quantitative analysis of the interaction of various factors, thereby accurately reflecting the overall impact of mining activities on the ecosystem. Through weight setting and weighted scoring, the model's ability to express multi-dimensional ecological damage is effectively improved, making damage assessment more scientific and accurate.

[0016] The present invention, by setting graded ecological damage thresholds and early warning mechanisms, can timely identify the damage level of the ecosystem, and automatically trigger corresponding early warnings and protection recommendations when the damage level threshold is reached or exceeded. This technology not only makes the response of ecological protection measures more rapid and more targeted, but also can effectively reduce the long-term and cumulative damage of deep-sea mining activities to the ecological environment, and ensure the sustainability of the marine ecosystem. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of an ecological damage assessment method according to an embodiment of the present invention; Figure 2 The figure is a schematic diagram of the process of performing wavelet decomposition on low-dimensional feature data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0020] like Figure 1-Figure 2 As shown, a method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis includes the following steps: S1: In the direct and indirect impact areas of mining disturbance, data on multiple disturbance factors are collected simultaneously, including physical data, chemical data, and biological data of water bodies and sediments to form an initial data set of multi-factor disturbance; S2: Preprocess the data of various disturbance factors collected by S1, extract the characteristic parameters of each disturbance factor through principal component analysis (PCA) and wavelet decomposition, and obtain the feature set of each factor to reduce the data dimension and highlight the main disturbance characteristics; S3: Based on the feature set extracted in S2, the association analysis algorithm is used to analyze the characteristic parameters of each disturbance factor, calculate the correlation between physical, chemical and biological factors, identify the combination of factors with interactions, and provide a basis for the construction of a multi-factor fusion model; S4: Based on the correlation analysis results of S3, a multi-factor integrated ecological damage assessment model is constructed to comprehensively assess the overall impact of various factor combinations on the ecosystem and generate ecological damage values; S5: Based on the ecological damage value generated by S4, set the damage threshold of the ecosystem to different combinations of disturbance factors and divide the ecological damage level; when the damage value reaches or exceeds the threshold of the corresponding level, trigger the corresponding early warning mechanism and generate corresponding ecological protection recommendations to prevent irreversible ecological damage caused by the superposition of multiple factors.

[0021] S1 specifically includes: S11: According to the geographical scope and equipment parameters of mining activities, several fixed sampling points are determined in the direct impact area and the indirect impact area, and the same sampling range is defined at each sampling point to ensure the coverage and consistency of data collection; based on the operating radius R of the mining equipment and the disturbance diffusion coefficient D, the formula is used Calculate the boundaries of the direct impact zone and the indirect impact zone, where t is the disturbance diffusion time; define the area within the operating radius R as the direct impact zone, where the disturbance caused directly by mining activities is the strongest; The area within the scope is defined as the indirect impact zone, which is indirectly affected by sediment spread and chemical dispersion; S12: At each sampling point in the direct impact zone and the indirect impact zone, use a multi-parameter water quality sensor to collect physical data of the water body, including temperature and turbidity, and use a sediment sampler to obtain physical data of the sediment to comprehensively record the physical disturbance characteristics; S13: Use dissolved oxygen monitors, conductivity sensors and metal ion concentration detection instruments to obtain chemical data of water and sediment at each sampling point, including dissolved oxygen concentration and concentration of target metal ions, to ensure that the chemical data fully records the impact of mining; S14: Collect benthic samples through benthic trawl sampling equipment, measure the benthic species and community density at each sampling point, and obtain biological data; S15: The collected physical, chemical and biological data are classified and summarized according to the sampling location and time sequence to generate a multi-factor disturbance initial data set covering the direct and indirect impact areas, providing basic data for subsequent analysis; by accurately dividing the direct and indirect impact areas and adopting corresponding data collection methods, a scientific initial data set is formed, which provides a solid data foundation for the comprehensive analysis of the ecological damage caused by mining disturbance, thereby improving the accuracy and scientificity of the ecological damage assessment.

[0022] S2 specifically includes: S21: Standardize the collected physical, chemical and biological data, and use data normalization or standardization algorithms to adjust each data item to the same level to eliminate the dimensional differences between different disturbance factors, making it easier to analyze later; S22: Perform principal component analysis (PCA) on the standardized physical, chemical and biological data, calculate the covariance matrix of each data item, and obtain the principal components through eigenvalue decomposition; select the first few principal components whose cumulative contribution rate reaches a preset threshold (such as 95%), and compress the original data into low-dimensional feature data containing the main feature information to reduce the data dimension and ensure the accuracy of data simplification; S23: performing wavelet decomposition on the low-dimensional feature data obtained in S22, and using the selected Daubechies wavelet to decompose the time series data into components of different scales; obtaining the variation characteristics of the data in different frequency bands through multi-scale analysis, and extracting characteristic parameters reflecting the disturbance intensity and frequency changes; S24: Screen the characteristic parameters extracted by principal component analysis and wavelet decomposition, retain the characteristic parameters of the disturbance factors that have the greatest impact on ecosystem damage, and group them into characteristic sets of physical, chemical and biological factors to represent the main characteristic information of each disturbance factor.

[0023] S23 specifically includes: S231: selecting Daubechies wavelet as the wavelet basis function and setting the scale layer number of wavelet decomposition to decompose the collected time series data; the time series data includes the measured values ​​of the physical, chemical or biological parameters of the mining disturbance area changing with time; since Daubechies wavelet has good localization characteristics, it can effectively capture the local change characteristics of the disturbance data; S232: Decomposing the time series data into multiple components of different scales; the decomposed data includes high-frequency detail components and low-frequency approximate components, wherein the high-frequency detail components are used to capture the instantaneous change characteristics of the disturbance, and the low-frequency approximate components are used to reflect the overall change trend; S233: Calculate the scale factor of each detail component to quantify the intensity of the disturbance at each scale; S234: Based on the scale factor of each detail component, by setting a threshold of the disturbance intensity, the component with significant disturbance intensity is extracted as a characteristic parameter; by setting the disturbance intensity threshold, the component with significant disturbance characteristics is extracted to ensure the pertinence and accuracy of data processing.

[0024] The specific calculation steps for wavelet decomposition of low-dimensional feature data are as follows: First, select Daubechies wavelet as the wavelet basis function, set the scale level J of wavelet decomposition (J represents the number of scale levels of decomposition), and express the time series data as ,in Represents the changes of physical, chemical or biological parameters in the mining disturbed area over time t; and Perform multi-scale wavelet decomposition and express it as components of different scales. The specific formula is: ,in, represents the detail component of the jth scale, j is the decomposition level index, Represents the approximate component of the Jth scale; detail component Capture the high-frequency disturbance characteristics of the data and approximate the components It reflects the low-frequency change trend of the data; Then, calculate each detail component The scale factor is used to measure the intensity of the disturbance at the jth scale; the scale factor is defined as: ,in, represents the scale factor of the j-th scale detail component, is the absolute value of the j-th scale detail component at time t. This scale factor is used to quantify the intensity of disturbances at each scale, which facilitates the subsequent analysis of the impact of each component. Finally, based on the detail components at each scale The scale factor is compared with the set disturbance intensity threshold T to extract the significant disturbance intensity component; by setting the disturbance intensity threshold , where T is the disturbance intensity threshold, is the empirical coefficient, max( ) is the maximum value of all detail component scale factors, which is used to screen significant disturbance features; select all scale factors that meet the conditions The amount of detail , and its scale factor and corresponding frequency information are used as the main characteristic parameters reflecting the disturbance intensity and frequency changes.

[0025] S24 specifically includes: S241: For each characteristic parameter extracted by principal component analysis and wavelet decomposition, calculate its influence on ecosystem damage. The influence is measured by the variance contribution rate of each characteristic parameter. The calculation formula is: ,in, represents the influence of the i-th feature parameter, represents the variance of the i-th feature parameter, is the total variance of all characteristic parameters; the variance contribution rate is used to quantify the explanatory power of each characteristic parameter in data changes. The greater the influence, the more significant the potential impact of the characteristic on ecosystem damage; S242: Based on the influence calculated in S241, all characteristic parameters are sorted by influence, and characteristic parameters whose influence reaches a preset threshold are retained. The preset threshold can be set by the cumulative contribution rate. For example, characteristic parameters with a cumulative influence of 95% are selected to ensure the explanatory power of characteristic parameters while reducing data redundancy. The selected characteristic parameters represent the factors that have the greatest impact on ecosystem damage; S243: The screened characteristic parameters are classified and aggregated according to physical, chemical and biological factors; for physical factor characteristic parameters, the aggregation includes water temperature, turbidity, etc.; for chemical factor characteristic parameters, the aggregation includes dissolved oxygen, metal ion concentration, etc.; for biological factor characteristic parameters, the aggregation includes benthic community density, species diversity, etc. The aggregated characteristic set can systematically reflect the ecological impact characteristics of different disturbance factors; by calculating the influence of characteristic parameters and screening out characteristic parameters with significant impact on ecosystem damage, a targeted physical, chemical and biological factor characteristic set is formed, which provides the most ecologically influential data support for the ecological damage assessment model, thereby improving the scientificity and accuracy of the ecological impact assessment.

[0026] S3 specifically includes: S31: Pair the physical, chemical and biological characteristic parameters in pairs and calculate the Pearson correlation coefficient between each pair of characteristic parameters to quantify the linear correlation between the two. The calculation formula of the Pearson correlation coefficient is: ,in, is the correlation coefficient between the characteristic parameters x and y, where x and y represent the values ​​of the two characteristic parameters respectively. and is the average value of the corresponding parameter; The value range is from -1 to 1. The closer the value is to 1 or -1, the stronger the linear correlation between the two parameters. S32: Perform a significance test on the calculated Pearson correlation coefficient to determine whether the correlation reaches a significant level; the significance test is determined by calculating the t value and comparing it with the set critical value level. The calculation formula of the t value is: , where t is the statistical value of the significance test, is the Pearson correlation coefficient between the two characteristic parameters, n is the number of samples, and the calculated t value is compared with the critical t value. If the absolute value of the t value is greater than the critical value, it is considered that the correlation reaches a significant level; S33: All pairs of characteristic parameters with significant correlation are used to form an association matrix. For all significantly correlated characteristic parameters, the Pearson correlation coefficient value of each pair of characteristic parameters is used as an element in the association matrix to form a matrix. The expression is: , where R is the incidence matrix, is the significant correlation coefficient between characteristic parameter i and characteristic parameter j; when When , the diagonal elements are 1, indicating the complete correlation of the parameters themselves; S34: Based on the correlation matrix, the characteristic parameter combinations with high correlation coefficient values ​​are screened, and the combinations with significant correlation among the physical, chemical and biological characteristic parameters are identified as the combinations of factors with interaction. Through this combination, the potential interaction relationship between physical, chemical and biological factors can be determined, providing an important basis for the subsequent multi-factor fusion ecological damage assessment model; the above steps accurately identify the combinations of factors with interaction through the correlation calculation and significance screening between the characteristic parameters, providing a scientific correlation basis for the construction of the multi-factor ecological damage model, and improving the explanatory power and accuracy of the model in ecological damage assessment.

[0027] S34 specifically includes: S341: To select the characteristic parameter combination with high correlation, set the threshold of correlation coefficient , usually between 0.7 and 1.0, is used to reflect strong positive or negative correlation; this threshold is used to determine which characteristic parameter combinations have significant interactions; S342: Traverse each element in the correlation matrix R and select all elements that meet the conditions The characteristic parameter combination of represents the correlation coefficient between characteristic parameter i and characteristic parameter j; when Greater than or equal to threshold When , the characteristic parameter pair is recorded and marked as a high correlation combination; S343: Classify the screened high-correlation characteristic parameter combinations according to three categories of physical, chemical and biological factors, assign each characteristic parameter to its corresponding factor category, and form a cross-category combination table to display the significant interaction relationship between physical, chemical and biological factors; by setting correlation thresholds, screening high-correlation parameter combinations and classifying them, the identification of significant interaction factors across categories is achieved, providing a scientific and reliable multi-factor combination basis for the ecological damage assessment model, thereby improving the accuracy and comprehensiveness of ecological damage analysis.

[0028] S4 specifically includes: S41: Based on the combination of highly correlated characteristic parameters identified in S3, determine the main input variables of physical, chemical and biological factors and construct the input variable set of the model; S42: Use the analytic hierarchy process to assign weights to each input variable and calculate the relative importance of each pair of factors in the ecological damage assessment by expert scoring or empirical weight setting; The specific steps are as follows: Construct a factor comparison matrix: Based on the sensitivity of the ecosystem to each factor, invite experts or use actual experience to compare the relative importance of each factor in pairs to generate a comparison matrix , where the elements Indicates the importance of factor i relative to factor j; if , then factor i is more important than factor j; if , then both are equally important; Standardization and weight calculation: Calculate the weight of each factor by standardizing the comparison matrix; sum the elements of each column in the comparison matrix, divide each element by the sum of its column, and get the standardized matrix B; the average value of each row is the weight of the corresponding factor , the calculation formula is: ,in, is the weight of the ith factor, is the element in the normalized matrix, is the number of factors; Consistency test: To ensure the consistency of the comparison matrix, calculate the consistency ratio. First, calculate the consistency index CI. The formula is: ,in, is the maximum eigenvalue of the matrix, and n is the order of the matrix; then the consistency ratio CR is calculated, the formula is: ,in, is the random consistency index. When the CR value is less than 0.1, the matrix meets the consistency requirements. Otherwise, the comparison matrix needs to be adjusted and recalculated.

[0029] S43: Based on the determined input variables and weights, a multi-factor fusion ecological damage assessment model is constructed, and the ecological damage value of each factor combination to the ecosystem is calculated; the above steps establish a comprehensive ecological damage assessment model through the reasonable allocation and weighted fusion of multi-factor weights, which can quantify the comprehensive impact of each factor combination on the ecosystem, generate reliable ecological damage values, and provide strong support for scientific assessment and decision-making of ecological damage.

[0030] The multi-factor fusion ecological damage assessment model constructed in S43 specifically includes: S431: Taking the determined physical, chemical and biological input variables as the input layer of the model, a multi-factor fusion ecological damage assessment model is constructed, and the expression is: , where D is the comprehensive ecological damage value of the ecosystem, is the weight of the ith factor, is the characteristic parameter value of the i-th factor, is the total number of factors; this model is used to comprehensively quantify the damage of multiple factors to the ecosystem; S432: multiplying the characteristic parameter value of each input variable by its corresponding weight to calculate the weighted damage contribution value of each factor; S433: To ensure the comparability of ecological damage values, the calculated damage values ​​are normalized to the range , which is convenient for comparison among multiple time periods or regions; the normalization formula is: ,in, is the normalized ecological damage value, and are the minimum and maximum values ​​of damage, respectively; through the weighted calculation and normalization processing of the multi-factor fusion ecological damage assessment model, the comprehensive impact of various factors on the ecosystem is accurately quantified, which improves the model's ability to express multi-dimensional ecological damage and provides a scientific basis for the quantitative assessment of ecological impacts.

[0031] S5 specifically includes: S51: Set the threshold of ecological damage classification based on historical ecological monitoring data and expert experience, and define corresponding thresholds for each damage level of the ecosystem; the damage threshold is set by using a hierarchical setting method, dividing the ecological damage value D into four levels, namely slight L1, moderate L2, severe L3 and extremely severe L4; the corresponding damage threshold is , , , and meet the following conditions: ; ; ; ;in, , , is the damage level threshold of the ecosystem; S52: Compare the ecological damage value generated by S4 with the classification threshold set in S51 to determine the current damage level of the ecosystem; according to the range of the damage value, divide it into corresponding damage levels L1, L2, L3 or L4 to determine the current ecological risk level; S54: When the calculated ecological damage value reaches or exceeds the set damage level threshold, the corresponding early warning mechanism is triggered and ecological protection recommendations are generated; the early warning mechanism will determine the warning level according to the current ecological damage level (L1 to L4), and the warning level corresponds to the ecological damage level, L1 to L4 correspond to low, medium, high and extremely high warnings respectively; protection recommendations include measures to slow down, suspend and repair the ecological environment of mining activities, and specific recommendations are generated based on the actual needs of each level to ensure the timeliness and pertinence of ecological protection measures; the above steps effectively monitor and respond to different degrees of ecological damage through the hierarchical setting of ecological damage thresholds and a multi-level early warning mechanism, ensuring that ecological protection recommendations can respond quickly to specific damage levels, thereby achieving effective management and control of ecological risks.

[0032] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0033] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis, characterized in that: The following steps are involved: S1: In the direct and indirect impact areas of mining disturbance, data on multiple disturbance factors are collected simultaneously, including physical data, chemical data, and biological data of water bodies and sediments to form an initial data set of multi-factor disturbance; S2: Preprocess the data of various disturbance factors collected by S1, extract the characteristic parameters of each disturbance factor by principal component analysis and wavelet decomposition, and obtain the characteristic set of each factor; S3: Based on the feature set extracted in S2, the association analysis algorithm is used to analyze the characteristic parameters of each disturbance factor, calculate the correlation between physical, chemical and biological factors, and identify the combination of factors with interactions; S4: Based on the correlation analysis results of S3, a multi-factor integrated ecological damage assessment model is constructed to comprehensively assess the overall impact of various factor combinations on the ecosystem and generate ecological damage values; S5: Based on the ecological damage value generated by S4, set the damage threshold of the ecosystem to different combinations of disturbance factors and divide the ecological damage level; when the damage value reaches or exceeds the threshold of the corresponding level, trigger the corresponding early warning mechanism and generate corresponding ecological protection recommendations.

2. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 1, characterized in that: The S1 specifically includes: S11: According to the geographical scope and equipment parameters of mining activities, determine a number of fixed sampling points in the direct impact area and the indirect impact area, and define the same sampling range at each sampling point to ensure the coverage and consistency of data collection; S12: At each sampling point in the direct impact zone and the indirect impact zone, use a multi-parameter water quality sensor to collect physical data of the water body, including temperature and turbidity, and use a sediment sampler to obtain physical data of the sediment to comprehensively record the physical disturbance characteristics; S13: Use dissolved oxygen monitors, conductivity sensors, and metal ion concentration detection instruments to obtain water and sediment chemical data at each sampling point, including dissolved oxygen concentration and target metal ion concentration; S14: Collect benthic samples through benthic trawl sampling equipment, measure the benthic species and community density at each sampling point, and obtain biological data; S15: The collected physical, chemical and biological data are classified and summarized according to the sampling location and time sequence to generate a multi-factor disturbance initial data set to provide basic data for subsequent analysis.

3. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 1, characterized in that: The S2 specifically includes: S21: Standardize the collected physical, chemical and biological data, and use data normalization or standardization algorithms to adjust each data item to the same order of magnitude to eliminate the dimensional differences between different disturbance factors; S22: Perform principal component analysis on the standardized physical, chemical and biological data, calculate the covariance matrix of each data item, and obtain the principal components through eigenvalue decomposition; and select the first few principal components whose cumulative contribution rate reaches the preset threshold, and compress the original data into low-dimensional feature data containing the main feature information to reduce the data dimension; S23: performing wavelet decomposition on the low-dimensional feature data obtained in S22, and using the selected Daubechies wavelet to decompose the time series data into components of different scales; obtaining the variation characteristics of the data in different frequency bands through multi-scale analysis, and extracting characteristic parameters reflecting the disturbance intensity and frequency changes; S24: Screen the characteristic parameters extracted by principal component analysis and wavelet decomposition, retain the characteristic parameters of the disturbance factors that have the greatest impact on ecosystem damage, and classify them into characteristic sets of physical, chemical and biological factors.

4. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 3 is characterized in that: The S23 specifically includes: S231: Select Daubechies wavelet as wavelet basis function and set the scale layer number of wavelet decomposition to decompose the collected time series data; S232: Decomposing the time series data into multiple components of different scales; the decomposed data includes high-frequency detail components and low-frequency approximate components, wherein the high-frequency detail components are used to capture the instantaneous change characteristics of the disturbance, and the low-frequency approximate components are used to reflect the overall change trend; S233: Calculate the scale factor of each detail component to quantify the intensity of the disturbance at each scale; S234: Based on the scale factor of each detail component, by setting a threshold of the disturbance intensity, a component with significant disturbance intensity is extracted as a feature parameter.

5. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 4 is characterized in that: The S24 specifically includes: S241: For each characteristic parameter extracted by principal component analysis and wavelet decomposition, calculate its influence on ecosystem damage. The influence is measured by the variance contribution rate of each characteristic parameter. The calculation formula is: ,in, represents the influence of the i-th feature parameter, represents the variance of the i-th feature parameter, is the total variance of all feature parameters; S242: based on the influence calculated in S241, all feature parameters are sorted by influence, and feature parameters whose influence exceeds a preset threshold are retained; S243: Classify and group the screened characteristic parameters according to physical, chemical and biological factors.

6. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 1, characterized in that: The S3 specifically includes: S31: Pair the physical, chemical and biological characteristic parameters in pairs and calculate the Pearson correlation coefficient between each pair of characteristic parameters to quantify the linear correlation between the two. The calculation formula of the Pearson correlation coefficient is: ,in, is the correlation coefficient between the characteristic parameters x and y, where x and y represent the values ​​of the two characteristic parameters respectively. and is the average value of the corresponding parameter; The value range is from -1 to 1. The closer the value is to 1 or -1, the stronger the linear correlation between the two parameters. S32: Perform a significance test on the calculated Pearson correlation coefficient to determine whether the correlation reaches a significant level; the significance test is determined by calculating the t value and comparing it with the set critical value level. The calculation formula of the t value is: , where t is the statistical value of the significance test, is the Pearson correlation coefficient between the two characteristic parameters, n is the number of samples, and the calculated t value is compared with the critical t value. If the absolute value of the t value is greater than the critical value, it is considered that the correlation reaches a significant level; S33: All pairs of characteristic parameters with significant correlation are used to form a correlation matrix, and the Pearson correlation coefficient value of each pair of characteristic parameters is used as an element in the correlation matrix to form a matrix, the expression is: , where R is the incidence matrix, is the significant correlation coefficient between characteristic parameter i and characteristic parameter j; when When , the diagonal elements are 1, indicating the complete correlation of the parameters themselves; S34: Based on the correlation matrix, characteristic parameter combinations with high correlation coefficient values ​​are screened, and combinations with significant correlation among physical, chemical and biological characteristic parameters are identified as combinations of factors with interaction.

7. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 6, characterized in that: The S34 specifically includes: S341: To select the characteristic parameter combination with high correlation, set the threshold of correlation coefficient ; S342: Traverse each element in the correlation matrix R and select all elements that meet the conditions The characteristic parameter combination of represents the correlation coefficient between characteristic parameter i and characteristic parameter j; when Greater than or equal to threshold When , the characteristic parameter pair is recorded and marked as a high correlation combination; S343: Classify the screened high-correlation characteristic parameter combinations according to the three factors of physics, chemistry and biology, assign each characteristic parameter to its corresponding factor category, and form a cross-category combination table.

8. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 1, characterized in that: The S4 specifically includes: S41: Based on the combination of highly correlated characteristic parameters identified in S3, determine the main input variables of physical, chemical and biological factors and construct the input variable set of the model; S42: Use the analytic hierarchy process to assign weights to each input variable and calculate the relative importance of each pair of factors in the ecological damage assessment by expert scoring or empirical weight setting; S43: Based on the determined input variables and weights, a multi-factor fusion ecological damage assessment model is constructed, and the ecological damage value of each factor combination to the ecosystem is calculated.

9. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 8, characterized in that: The S43 specifically includes: S431: Taking the determined physical, chemical and biological input variables as the input layer of the model, a multi-factor fusion ecological damage assessment model is constructed, and the expression is: , where D is the comprehensive ecological damage value of the ecosystem, is the weight of the ith factor, is the characteristic parameter value of the i-th factor, is the total number of factors; S432: multiplying the characteristic parameter value of each input variable by its corresponding weight to calculate the weighted damage contribution value of each factor; S433: To ensure the comparability of ecological damage values, the calculated damage values ​​are normalized to the range .

10. The method for evaluating ecological damage caused by deep-sea mining disturbance based on data analysis according to claim 9, characterized in that: The S5 specifically includes: S51: Set the threshold of ecological damage classification based on historical ecological monitoring data and expert experience, and define corresponding thresholds for each damage level of the ecosystem; the damage threshold is set by using a hierarchical setting method, dividing the ecological damage value D into four levels, namely slight L1, moderate L2, severe L3 and extremely severe L4; the corresponding damage threshold is , , , and meet the following conditions: ; ; ; ;in, , , is the damage level threshold of the ecosystem; S52: Compare the ecological damage value generated by S4 with the classification threshold set in S51 to determine the current damage level of the ecosystem; according to the range of the damage value, divide it into corresponding damage levels L1, L2, L3 or L4 to determine the current ecological risk level; S54: When the calculated ecological damage value reaches or exceeds the set damage level threshold, the corresponding early warning mechanism is triggered and ecological protection recommendations are generated; the early warning mechanism will determine the early warning level according to the current ecological damage level, and the early warning level corresponds to the ecological damage level, L1 to L4 correspond to low, medium, high and extremely high warnings respectively; the protection recommendations include measures to slow down, suspend and repair the ecological environment of mining activities.

Images