Quantitative analysis method and system for spatial differentiation characteristics of water quality in water area

Abstract

The invention belongs to but not limited to the technical field of water quality analysis, and particularly relates to a water area water quality spatial differentiation characteristic quantitative analysis method and system. Comprehensively evaluating the water quality by adopting the water quality index; evaluating the eutrophication state of the water area by adopting the comprehensive nutrition state index Analyzing water quality spatial distribution characteristics by using a self-organizing neural network; and analyzing whether the water quality has spatial correlation and correlation degree through spatial autocorrelation. According to the technical scheme, the water quality condition of the water area can be determined, and main pollution indexes are identified; water area water quality space distribution characteristics can be revealed, and an important basis is provided for precise pollution control.

Description

Technical Field

[0001] The present invention belongs to but is not limited to the field of water quality analysis technology, and in particular relates to a method and system for quantitatively analyzing spatial differentiation characteristics of water quality in an aquatic area. Background Art

[0002] Lakes, as precious surface water resources, are the lifeline of human social and economic development. With the intensification of global climate change and human activities, lake water safety is facing unprecedented threats. Therefore, clarifying the water quality of lakes and the risk of blue algae blooms and proposing strategies for lake water environment protection are urgent issues to ensure water supply security and regional social and economic development. However, as important sources of water supply, lakes often have a large water area and their water quality spatial distribution shows significant differences. How to quantitatively analyze pollution patterns, accurately identify important pollution areas, and accurately implement policies are the problems faced by lake water pollution control.

[0003] In view of the above analysis, the technical problems that need to be solved urgently in the prior art are:

[0004] 1) The singularity and locality of water quality data:

[0005] In existing technologies, water quality monitoring is usually limited to single indicator analysis or local area data, which cannot fully reflect the overall water quality status and spatial distribution characteristics of the water area, resulting in a lack of scientific basis for governance measures.

[0006] 2) Lack of systematic identification of major pollution factors:

[0007] In traditional water quality analysis, the identification of pollution factors mostly uses empirical methods or simple statistical means, which makes it difficult to reveal the deep-seated driving factors of complex water quality problems.

[0008] 3) The spatial distribution of water quality is unclear:

[0009] Existing water quality research often ignores the spatial distribution differences of water quality in water bodies, and is unable to identify concentrated areas and diffusion trends of pollution, resulting in a lack of targeted governance measures.

[0010] 4) Insufficient research on spatial correlation:

[0011] Water bodies are generally large in area, and there is a potential interdependence between their water quality conditions and the locations of sampling points. Traditional analysis methods lack research on the spatial correlation of water quality and are unable to identify the impact of water quality between adjacent areas. Summary of the invention

[0012] In view of the problems existing in the prior art, the present invention provides a method and system for quantitatively analyzing the spatial differentiation characteristics of water quality in water areas.

[0013] The present invention is achieved by a method for quantitatively analyzing the spatial differentiation characteristics of water quality in a water area, comprising:

[0014] S1, collect water samples from water areas regularly and analyze and test water quality indicators;

[0015] S2, using a comprehensive assessment method to assess water quality and eutrophication status in water bodies;

[0016] S3, factor identification analysis of water quality;

[0017] S4, using self-organizing neural network to analyze the spatial distribution characteristics of water quality;

[0018] S5, analyze whether there is spatial correlation in water quality and the degree of correlation through spatial autocorrelation.

[0019] Furthermore, the comprehensive assessment methods include: water quality index and comprehensive trophic status index;

[0020] The water quality index is used to comprehensively evaluate water quality, and its calculation formula is as follows:

[0021]

[0022] Where WQI is the comprehensive water quality index, Ci is the standardized score of water quality factor i, Pi is the weight of water quality factor i, the minimum value of Pi is 1, and the maximum value is 4; the WQI evaluation value range is 0-100, and the larger the value, the higher the water quality health; according to the score, water quality is divided into 5 levels: excellent [90-100], good [70-90], general [50-70], poor [25-50], and very poor [0-25).

[0023] Furthermore, the comprehensive trophic status index is used to evaluate the eutrophication status of water bodies: the evaluation factors include ρ(Chla), ρ(TP), ρ(TN), SD, and ρ(CODMn); ​​taking ρ(Chla) as the benchmark parameter, the normalized relevant weight calculation formula of the jth parameter is:

[0024]

[0025] Where rij is the correlation coefficient between the jth parameter and the benchmark parameter p(Chla), and m is the number of evaluation parameters;

[0026] The calculation formulas of nutritional status index of each parameter are as follows: TLI(Chla)=10(2.5+1.086×ln[Chla]); TLI(TP)=10(9.436+1.624×ln[TP]); TLI(TN)=10(5.453+1.694×ln[TN]); TLI(SD)=10(5.118-1.94×ln[SD]); TLI(CODMn)=10(0.109+2.661×ln[CODMn]);

[0027] Wherein: TLI(Chla), TLI(TP), TLI(TN), TLI(SD), TLI(CODMn) are the comprehensive nutrient status indexes of Chla, TP, TN, SD, CODMn respectively; [Chla] is ρ(Chla), mg / m3; [TP], [TN], [CODMn] are ρ(TP), ρ(TN), ρ(CODMn), mg / L respectively; [SD] is SD, m; According to the score, the eutrophication status is divided into 5 levels: oligotrophic (0-30], mesotrophic [30-50], slightly eutrophic (50-60], moderately eutrophic (60-70], and severely eutrophic (70-100).

[0028] Conduct factor identification analysis on water quality, including:

[0029] The three common factors of nitrogen and phosphorus indexes, physical indexes, and organic matter and eutrophication indexes were extracted to calculate the variance contribution rate.

[0030] Furthermore, the spatial distribution characteristics of water quality were analyzed using self-organizing neural networks as follows:

[0031] By inputting the tested water quality sample data, the neural network will self-organize and automatically classify the training samples. The specific implementation process includes:

[0032] Initialization of SOM model parameters: Initially assign a small random weight vector W = (Wi) to each neural unit, i = 1, 2, ..., m, where Wi = (ω1, ω2, ..., ωp); determine the initial value Ng(0) of the neighborhood Ng(t), and the learning rate is η; input the water quality sample feature data set X = (Xj), j = 1, 2, ...n, where Xj = (x1, x2, ..., xq);

[0033] ② Vector normalization: Normalize the weight vector and sample vector to obtain Wi′=(ω1′,ω2′,…,ωp′) and Xj′=(x1′,x2′,…,xq′) respectively;

[0034] ③ Find the winning neuron g: Find the minimum distance between the input sample vector Xj′ and the weight vector Wi′ corresponding to each neuron in the competition layer, and calculate the minimum Euclidean distance d between the two; dg = min[dj], j = 1, 2, ... m;

[0035] ④ Adjust network output and weights: update weight Wj(t+1)=Wj(t)+α[Xj-Wj(t)], neighborhood Ng(t)=int{Ng(0)}[1-t / T];

[0036] ⑤ Repeat steps 3 and 4 until the number of learning times reaches T.

[0037] Spatial autocorrelation is used to analyze whether there is spatial correlation and the degree of correlation in water quality. The degree of correlation is reflected by global Moran's I, and the calculation formula is as follows:

[0038]

[0039] Where: n is the number of spatial regions studied; Xi and Xj are the water quality of regions i and j respectively, and is the mean water quality of the entire study area; Wij is the spatial weight matrix;

[0040] Moran's I is between -1 and 1. The closer the absolute value is to 1, the stronger the spatial autocorrelation is. A positive value indicates that Xi and Xj change in the same direction and the data are positively correlated. A negative value indicates that Xi and Xj change in opposite directions and the data are negatively correlated. A value of 0 indicates that the data is randomly distributed and has no autocorrelation.

[0041] Furthermore, the data is searched for missing values, error values ​​are eliminated, and standardized to remove dimensions. Specifically:

[0042] Before data analysis, SPSS19 software was used to find missing values, eliminate error values, and standardize the data. The KS normality test of water quality data was performed at a significance level of 0.05.

[0043] Based on SPSS19, the spatial characteristics and main pollution factors of lake water quality were analyzed by factor analysis, and the KMO and spherical Bartlett test methods were used.

[0044] Furthermore, water samples are collected regularly and water quality indicators are analyzed and tested, specifically:

[0045] Nine typical water quality indicators, including pH, ρ(DO), ρ(TP), ρ(TN), ρ(NH3-N), ρ(CODMn), ρ(SS), ρ(NO3-N), and ρ(Chla), were selected for analysis. Among them, pH and ρ(DO) were measured on-site using a portable dissolved oxygen meter, and other indicators were measured by refrigerating water samples and then bringing them back to the laboratory.

[0046] Another object of the present invention is to provide a system for quantitatively analyzing spatial differentiation characteristics of water quality in a water area, which implements the method for quantitatively analyzing spatial differentiation characteristics of water quality in a water area, comprising:

[0047] Analysis and testing module, regularly collects water samples from water areas and analyzes and tests water quality indicators;

[0048] Comprehensive assessment module, which uses comprehensive assessment methods to assess water quality and eutrophication status of water bodies;

[0049] Factor identification and analysis module, which performs factor identification and analysis on water quality;

[0050] Spatial distribution characteristics analysis module, using self-organizing neural network to analyze the spatial distribution characteristics of water quality;

[0051] The correlation analysis module uses spatial autocorrelation to analyze whether there is spatial correlation in water quality and the degree of correlation.

[0052] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for quantitatively analyzing the spatial differentiation characteristics of water quality in a water area.

[0053] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the method for quantitatively analyzing the spatial differentiation characteristics of water quality in a water area.

[0054] Another object of the present invention is to provide an information data processing terminal, which includes the quantitative analysis system for spatial differentiation characteristics of water quality in water areas.

[0055] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0056] First, the machine learning method for analyzing the spatial differentiation characteristics of water quality provided by the present invention is:

[0057] By regularly collecting water samples from water areas and analyzing water quality indicators, the real-time and authenticity of water quality indicators are guaranteed.

[0058] The use of water quality index to comprehensively evaluate water quality can analyze the true situation of water quality areas.

[0059] The comprehensive nutrient status index is used to evaluate the eutrophication status of water bodies, which can obtain the true and accurate eutrophication status of water bodies.

[0060] By identifying and analyzing water quality factors, we can accurately obtain water pollution factors. We can use self-organizing neural networks to analyze the spatial distribution characteristics of water quality; we can obtain water quality characteristics.

[0061] The spatial distribution of .

[0062] Spatial autocorrelation is used to analyze whether there is spatial correlation in water quality and the degree of correlation; it is possible to determine whether there is spatial correlation in water quality parameters and the specific degree of correlation.

[0063] Through the above technical scheme, the present invention can clarify the water quality status of the water area and identify the main pollution indicators; it can reveal the temporal and spatial variation patterns of the water quality of the water area; and by establishing a quantitative relationship between the eutrophication status of the lake and the main pollution indicators, it can ensure accurate analysis of the water quality of the water area.

[0064] Second, the technical solution of the present invention quantifies the spatial correlation of water quality through spatial autocorrelation analysis, providing a decision-making basis for regional joint prevention and control of pollution.

[0065] Significant technological progress achieved by this technical solution in industrial applications:

[0066] 1) Improved data comprehensiveness and accuracy:

[0067] By regularly collecting full-coverage water sample data and combining comprehensive evaluation methods and multivariate statistical techniques, a comprehensive analysis of water quality indicators can be achieved, thereby improving the scientificity and accuracy of water quality assessment.

[0068] 2) Scientific identification of pollution factors:

[0069] By using factor identification methods such as principal component analysis, the main influencing factors are extracted from complex water quality data, the sources of pollution are identified, and a scientific basis is provided for precise governance, which significantly improves the efficiency and reliability of pollution tracing.

[0070] 3) Accurately reveal spatial distribution characteristics:

[0071] Using self-organizing neural network technology, water quality data is mapped from multidimensional space to a two-dimensional clustering diagram, clearly revealing the water quality distribution characteristics and differences in different regions, and significantly improving the spatial analysis capabilities of water areas.

[0072] 4) Quantitative analysis of spatial correlation:

[0073] Through the spatial autocorrelation analysis of the global and local Moran index, the spatial aggregation degree and hot spots of water quality are quantified, which solves the problem of lack of spatial correlation analysis in the past and provides an important reference for joint prevention and control.

[0074] 5) Wide adaptability of application scenarios:

[0075] This technical solution is applicable to various types of water bodies (such as lakes, rivers, and wetlands), and has broad application prospects in the fields of water resources management, pollution prevention and control, and ecological restoration, providing systematic and scientific technical support for water environment management.

[0076] 6) Improved targeting and effectiveness of governance measures:

[0077] By revealing the spatial distribution patterns of water quality and the main pollution factors, this technical solution can formulate differentiated governance measures for different regions, effectively avoid waste of resources and governance blind spots, and greatly improve the efficiency and effectiveness of water environment governance.

[0078] This technical solution achieves accurate analysis of the spatial differentiation characteristics of water quality through the integrated application of multiple technologies (such as principal component analysis, self-organizing neural network and spatial autocorrelation analysis). Compared with existing technologies, its significant technical progress is reflected in:

[0079] the comprehensiveness of data collection and evaluation;

[0080] The scientific nature of pollution factor identification;

[0081] Accurately reveal the spatial distribution of water quality;

[0082] Innovation in quantifying spatial correlation.

[0083] This plan provides a more efficient, intelligent and scientific solution for aquatic ecological protection and pollution control, and has important industrial application value and promotion potential. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is a flow chart of a machine learning method for analyzing water quality spatial differentiation characteristics provided by an embodiment of the present invention.

[0085] Figure 2 This is a schematic diagram of the lake location and sampling point layout provided by an embodiment of the present invention.

[0086] Figure 3 Schematic diagram of the distribution of the water quality comprehensive index (a) and the eutrophication comprehensive index (b) provided in the embodiment of the present invention

[0087] Figure 4 Schematic diagram of factor scores of water quality rotation space at different time and space scales provided by an embodiment of the present invention

[0088] Figure 5 Schematic diagram of the analysis result of the self-organizing neural network provided by the embodiment of the present invention

[0089] Figure 6 Schematic diagram of the Spearman correlation coefficient between lake water quality indicators and TLI at different spatial scales provided by the embodiment of the present invention

[0090] Figure 7 It is a structural diagram of a machine learning system for analyzing spatial differentiation characteristics of water quality provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0091] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0092] The quantitative analysis method of the spatial heterogeneity of water quality in water areas obtains water quality data covering the entire study area by regularly collecting water samples in the water area. The collection points are reasonably arranged according to the functional area distribution of the water area and the possible impact range of the pollution source to ensure that the data is representative. The water samples are analyzed in the laboratory to test conventional water quality indicators such as chemical oxygen demand (COD), total nitrogen (TN), total phosphorus (TP), dissolved oxygen (DO) and pH value. The data results provide basic input for subsequent evaluation.

[0093] Using comprehensive assessment methods (such as the pollution index method or the comprehensive water quality index method), the water quality indicators detected are weighted to obtain the comprehensive water quality status of the water area. For the assessment of eutrophication, the eutrophication index (TLI) is calculated based on indicators such as total nitrogen, total phosphorus, and chlorophyll-a, and different eutrophication levels (such as mild, moderate, and severe) are divided. This step clarifies the overall water quality status of the water area and the spatial distribution of eutrophication, providing direction for subsequent factor identification.

[0094] Multivariate statistical methods (such as principal component analysis or factor analysis) are used to reduce the dimensionality of water quality data and identify the factors that mainly affect water quality. For example, principal component analysis reduces the dimensions of complex indicators by extracting the main variance contributing factors in the data, and attributes water quality problems to several main driving factors (such as agricultural non-point source pollution, industrial emissions, natural background concentrations, etc.). This process can provide a scientific basis for pollution tracing and management.

[0095] The water quality data was clustered using the Self-Organizing Map (SOM) algorithm. SOM uses unsupervised learning methods to map water quality data with multidimensional characteristics into a two-dimensional grid and classify similar water quality samples into the same cluster. This process can intuitively display the spatial distribution characteristics of water quality in different regions and reveal the potential patterns of water pollution and their local differences.

[0096] Spatial statistical methods (such as the global Moran's index and the local Moran's index) are used to conduct correlation analysis on the spatial distribution of water quality. The global Moran's index is used to quantify whether there is overall spatial correlation in water quality (such as positive correlation indicates that highly polluted areas are concentrated, and negative correlation indicates that polluted areas are dispersed); the local Moran's index reveals the aggregation effect of specific areas, identifying hot spots with significantly similar water quality and isolated points with significantly different water quality. This analysis provides a scientific basis for spatial management and zoning governance.

[0097] Combining eutrophication status assessment, factor identification, self-organizing neural network and spatial autocorrelation results, the spatial differentiation characteristics of water quality are comprehensively analyzed. By visualizing the analysis results, the pollution sources, spatial distribution patterns and impact levels are displayed. Ultimately, data support and scientific basis are provided for the precise management of water areas, pollution prevention and control, and the formulation of ecological restoration measures.

[0098] This method integrates laboratory testing, statistical analysis and machine learning technology to deeply analyze the spatial differentiation characteristics of water quality in aquatic areas in a multi-dimensional way. By accurately identifying the main pollution factors, revealing the spatial distribution patterns and regional correlations, it provides quantitative support for water environment management and has high practicality and promotion value.

[0099] Embodiment 1:

[0100] This embodiment will take the typical shallow lakes in the Huaihe River Basin as the research object to illustrate the technical solution of the present invention in detail.

[0101] Lakes, as precious surface water resources, are the lifeline of human social and economic development. With the intensification of global climate change and human activities, lake water safety is facing unprecedented threats. According to statistics, in 2020, among the 112 important lakes and reservoirs in my country where water quality monitoring was carried out, the proportion of water pollution (Class III and above) reached 23.2%, and the proportion of eutrophication reached 29%. Therefore, clarifying the spatial variation characteristics of lake water quality and the risk of cyanobacterial blooms is an important issue that needs to be urgently addressed in the current lake water environment field, which is of great significance to ensuring lake water safety.

[0102] Therefore, clarifying the water quality of lakes and the risk of blue algae blooms, and proposing strategies for lake water environment protection are urgent issues to ensure the water quality safety of the Huaihe River Basin and regional social and economic development. Based on this, the main research objectives of this invention include clarifying the water quality of lakes, identifying major pollution indicators, revealing the spatial variation of lake water quality, and providing technical support for local precision pollution control, water quality safety of shallow lakes in the Huaihe River Basin, and protection and management of similar lakes.

[0103] The specific plan is as follows:

[0104] 1. Materials and methods

[0105] (1) Sample collection and analysis:

[0106] 32 surface water samples were collected from a typical shallow lake in the Huaihe River Basin.

[0107] The sampling points are collected once a month. The locations and numbers of sampling points are determined according to the "HJ / T 91-2002 Technical Specifications for Surface Water and Sewage Monitoring" and "Technical Guidelines for Lake Ecological Safety Investigation and Assessment". The points are arranged according to the grid method, such as Figure 2 Nine typical water quality indicators, including pH, ρ(DO), ρ(TP), ρ(TN), ρ(NH3-N), ρ(CODMn), ρ(SS), ρ(NO3-N), and ρ(Chla), were selected for analysis. pH and ρ(DO) were measured on-site using a portable dissolved oxygen meter (HQ40d, HACH, USA), and other indicators were measured by refrigerating water samples and bringing them back to the laboratory.

[0108] (2) Research methods

[0109] The water quality index (WQI) is used to comprehensively evaluate the lake water quality, and its calculation formula is as follows:

[0110]

[0111] Where WQI is the comprehensive water quality index, Ci is the standardized score of water quality factor i, Pi is the weight of water quality factor i, the minimum value of Pi is 1, and the maximum value is 4; the WQI evaluation value range is 0-100, and the larger the value, the higher the water quality health; according to the score, water quality is divided into 5 levels: excellent [90-100], good [70-90], general [50-70], poor [25-50], and very poor [0-25).

[0112] The comprehensive trophic state index (TLI) is used to evaluate the eutrophication status of water bodies. The evaluation factors include ρ(Chla), ρ(TP), ρ(TN), SD, and ρ(CODMn). Taking ρ(Chla) as the benchmark parameter, the normalized weight calculation formula of the jth parameter is:

[0113]

[0114] Wherein, rij is the correlation coefficient between the jth parameter and the reference parameter p(Chla), and m is the number of evaluation parameters. In this embodiment, m is 32.

[0115] The calculation formulas of nutritional status index of each parameter are as follows: TLI(Chla)=10(2.5+1.086×ln[Chla]); TLI(TP)=10(9.436+1.624×ln[TP]); TLI(TN)=10(5.453+1.694×ln[TN]); TLI(SD)=10(5.118-1.94×ln[SD]); TLI(CODMn)=10(0.109+2.661×ln[CODMn]);

[0116] Wherein: TLI(Chla), TLI(TP), TLI(TN), TLI(SD), TLI(CODMn) are the comprehensive nutrient status indexes of Chla, TP, TN, SD, CODMn respectively; [Chla] is ρ(Chla), mg / m3; [TP], [TN], [CODMn] are ρ(TP), ρ(TN), ρ(CODMn), mg / L respectively; [SD] is SD, m; According to the score, the eutrophication status is divided into 5 levels: oligotrophic (0-30], mesotrophic [30-50], slightly eutrophic (50-60], moderately eutrophic (60-70], and severely eutrophic (70-100).

[0117] Use self-organizing neural network to analyze the spatial distribution characteristics of water quality.

[0118] The self-organizing map (SOM) is a clustering unsupervised learning algorithm. It is an artificial neural network that simulates the signal processing characteristics of the human brain. It is mainly composed of an input layer and a competition layer. By inputting the tested water quality sample data, the neural network will self-organize and map, and automatically classify the training samples. The specific implementation process includes:

[0119] 1) Initialization of SOM model parameters: Initially assign a small random weight vector W = (Wi) to each neural unit, i = 1, 2, ..., m, where Wi = (ω1, ω2, ..., ωp). Determine the initial value Ng(0) of the neighborhood Ng(t) and the learning rate η. Input the sample feature data set X = (Xj), j = 1, 2, ...n, where Xj = (x1, x2, ..., xq).

[0120] 2) Vector normalization: Normalize the weight vector and sample vector respectively to obtain Wi′=(ω1′, ω2′, …, ωp′) and Xj′=(x1′, x2′, …, xq′).

[0121] 3) Find the winning neuron g: Find the minimum distance between the input sample vector Xj′ and the weight vector Wi′ corresponding to each neuron in the competition layer, and calculate the minimum Euclidean distance d between the two.

[0122] dg=min[dj], j=1,2,...m.

[0123] 4) Adjust network output and weights: update weight Wj(t+1)=Wj(t)+α[Xj-Wj(t)], neighborhood Ng(t)=int{Ng(0)}[1-t / T];

[0124] 5) Repeat steps 3 and 4 until the number of learning times reaches T.

[0125] To determine whether water quality parameters have spatial correlation and the degree of correlation, this experiment used spatial autocorrelation to analyze whether there is spatial correlation in water quality parameters and the degree of correlation.

[0126] The example is analyzed by spatial autocorrelation, and the degree of correlation is reflected by global Moran's I.

[0127] The calculation formula is as follows:

[0128]

[0129] Where: n is the number of spatial regions studied; Xi and Xj are the water quality in regions i and j respectively, and is the mean water quality of the entire study area; Wij is the spatial weight matrix. Moran's I is between -1 and 1. The closer the absolute value is to 1, the stronger the spatial autocorrelation is.

[0130] A positive value indicates that Xi and Xj change in the same direction and the data is positively correlated; a negative value indicates that Xi and Xj change in opposite directions and the data is negatively correlated; a value of 0 indicates that the data is randomly distributed and has no autocorrelation.

[0131] 2. Results and analysis:

[0132] (1) Overall characteristics of water quality

[0133] The overall situation of lake water quality is shown in Table 1. The ranges of conventional indicators pH, ρ(SS), and ρ(DO) are 7.06-8.96 and 1-38 mg respectively. Before data analysis, SPSS19 software was used to find missing values, exclude error values, and standardize the data. At a significance level of 0.05, the KS (Kolmogorov-Smirnov) normality test of water quality data was performed. Due to the non-normality of the data distribution in this article, the Spearman rank correlation test was used to analyze the temporal trend of lake water quality and the correlation of water quality indicators, and the Kruskal-Wallis non-parametric test was used to analyze the seasonal differences in lake water quality. At the same time, based on SPSS19, the spatial characteristics and main pollution factors of lake water quality were analyzed by factor analysis, and the analysis method used was KMO and spherical Bartlett test.

[0134] Table 1 Statistical analysis of water quality indicators

[0135]

[0136] (2) Lake water pollution status

[0137] ①Water quality and eutrophication assessment

[0138] The evaluation found that Figure 3 As shown in the figure, the comprehensive water quality index WQI of the lake water body is 66.42-73.18, and the water quality is in general and good state. Spatially, the WQI value in the southwest of the lake is higher, and the water quality is good and in good state. The WQI value in the area where the Yi River and the Central Canal enter the lake is lower, and the water quality is relatively poor. The comprehensive eutrophication index TLI of the lake is 44.99-52.89, and the overall nutrient state is in a medium nutrient and slightly eutrophic state. The TLI value of the area where the Central Canal enters the lake, the southwest and northwest of the lake is lower, and it is in a medium nutrient state. The TLI values ​​of other areas are all higher, and the overall state is slightly eutrophic.

[0139] ②Identification of major water pollution factors

[0140] Three common factors were extracted through factor analysis, and the cumulative variance contribution rate was 64.37%. From the rotation component matrix (Table 2), it can be seen that the first common factor is mainly nitrogen and phosphorus indicators, with a variance contribution rate of 29%; the second common factor is mainly physical indicators, with a variance contribution rate of 20%; the third common factor is mainly organic matter and eutrophication indicators, with a variance contribution rate of 15%.

[0141] Table 2 Rotated component matrix

[0142]

[0143] Factor score rotation plot ( Figure 4) shows that the first common factor scores of points 3, 5, 8, and 9 are high, and nitrogen and phosphorus pollution is serious. The second common factor scores of points 17, 20, 21, and 31 are high, and the organic matter and eutrophication indicators are high. In August 2019 and August 2020, the water quality was seriously polluted, and the scores of the first two common factors were high.

[0144] (3) Spatial distribution characteristics of lake water quality

[0145] After calculation, the size of the self-organizing neural network is 32 (8×4), QE and TE both meet the SOM network training conditions (QE=1.645, TE=0.000), and the optimal number of clusters is 4, that is, the lake is spatially divided into four clustering spaces ( Figure 5 c) Spatial component diagram of water quality indicators Figure 5 As shown, the spatial variation characteristics of ρ(TN), ρ(TP), ρ(NO3-N), and ρ(NH3-N) are relatively consistent, all showing a characteristic of gradually decreasing concentration from northeast to southwest, ρ(CODMn) decreases from east to west, ρ(Chla) and TLI decrease from southeast to northwest, ρ(SS) and pH decrease from southwest to northeast, and ρ(DO) decreases from south to north.

[0146] Through the global Moran index analysis (Table 3), it was found that except for ρ(CODMn) and ρ(NH3-N) whose P values ​​were greater than the significance level of 0.05, the other indicators of lake water quality were all less than the significance level of 0.01 or 0.05, indicating that there were spatial differences in ρ(CODMn) and ρ(NH3-N), and there was spatial autocorrelation in other indicators.

[0147] Table 3 Global Moran index values ​​of lake water quality indicators

[0148]

[0149] 3. Conclusion:

[0150] (1) The lake water quality index (WQI) was 66.42-73.18, indicating that the water quality was in a general or good state. The lake eutrophication comprehensive index (TLI) was 44.99-52.89, indicating that the nutrient status was in a medium or slightly eutrophic state.

[0151] (2) The indicators that play a leading role in lake water quality are nitrogen and phosphorus indicators, organic matter and eutrophication indicators, and physical indicators. The variance contribution rates are 29%, 15%, and 20%, respectively.

[0152] (3) In terms of spatial variation, the spatial variation characteristics of nitrogen and phosphorus indicators are relatively consistent, showing high concentrations in the northeast and low concentrations in the southwest. Organic matter and eutrophication indicators show high concentrations in the east and low concentrations in the west. Physical indicators show high concentrations in the south and low concentrations in the north. In addition, except for ρ(CODMn),

[0153] Except ρ(NH3-N), other water quality indicators have spatial autocorrelation.

[0154] like Figure 7 As shown in the figure, the quantitative analysis system of spatial differentiation characteristics of water quality in water areas includes:

[0155] Analysis and testing module, regularly collects water samples from water areas and analyzes and tests water quality indicators;

[0156] Comprehensive assessment module, which uses comprehensive assessment methods to assess water quality and eutrophication status of water bodies;

[0157] Factor identification and analysis module, which performs factor identification and analysis on water quality;

[0158] Spatial distribution characteristics analysis module, using self-organizing neural network to analyze the spatial distribution characteristics of water quality;

[0159] The correlation analysis module uses spatial autocorrelation to analyze whether there is spatial correlation in water quality and the degree of correlation.

[0160] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0161] The present invention uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the method and core ideas of the present invention. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

[0162] An application embodiment of the present invention provides a computer device, which includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor executes the steps of a method for quantitatively analyzing spatial differentiation characteristics of water quality in a water area.

[0163] An application embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of a method for quantitatively analyzing spatial differentiation characteristics of water quality in a water area.

[0164] An application embodiment of the present invention provides an information data processing terminal, which includes a quantitative analysis system for spatial differentiation characteristics of water quality in a water area.

[0165] Example 1: Application in lake eutrophication control

[0166] A large lake suffered from severe eutrophication due to agricultural non-point source pollution and urban drainage, resulting in algal bloom, which had a significant impact on the lake ecosystem and the lives of surrounding residents.

[0167] 1. Water quality data collection and index analysis:

[0168] Sampling points are set up in different functional areas of the lake (such as agricultural drainage areas, urban sewage areas and water source protection areas), and water samples are collected at regular intervals to test key water quality indicators such as total nitrogen (TN), total phosphorus (TP), chlorophyll-a, and dissolved oxygen (DO).

[0169] The eutrophication status of each area is assessed and divided into light, moderate and severe eutrophication areas.

[0170] 2. Identification of pollution factors:

[0171] Using principal component analysis techniques, agricultural non-point source pollution (such as nitrogen and phosphorus loss) and urban drainage (organic pollutants) were identified as the main driving factors of lake eutrophication.

[0172] 3. Analysis of spatial distribution characteristics:

[0173] A cluster analysis of water quality data from different sampling points using a self-organizing neural network (SOM) revealed that the total nitrogen and total phosphorus concentrations were highest in agricultural drainage areas, while the chlorophyll-a value in urban sewage areas was significantly increased.

[0174] 4. Spatial autocorrelation analysis:

[0175] Through the calculation of the Moran Index, it was identified that the high nitrogen and high phosphorus concentrations caused by agricultural non-point source pollution formed a significant spatial cluster in the downstream area of ​​the lake.

[0176] Application effect:

[0177] Accurately identify pollution sources: It is clear that agricultural drainage is the main source of pollution, and urban sewage is a secondary factor.

[0178] Differentiated governance: Promote buffer zone planting and fertilization optimization measures in agricultural areas, and strengthen pollution control and ecological restoration in urban areas.

[0179] Improvement of eutrophication: After one year of treatment, the total nitrogen and total phosphorus concentrations in the lake decreased by more than 20%, the chlorophyll-a concentration decreased by 15%, and the algal bloom phenomenon was significantly reduced.

[0180] Example 2: Application in urban river water quality zoning management

[0181] Due to the influence of industrial emissions and domestic sewage, the overall water quality of a city's river has deteriorated, COD in some river sections has exceeded the standard, and the ecosystem is facing serious threats.

[0182] Application Process:

[0183] 1. Water quality data collection and index analysis:

[0184] Multiple sampling points were set up in the upper, middle and lower reaches of the river to detect COD, ammonia nitrogen (NH4 + -N), total organic carbon (TOC) and other key indicators to obtain water quality data for each river section.

[0185] 2. Identification of pollution factors:

[0186] Through factor analysis, it was identified that industrial wastewater discharge was the main source of upstream pollution, while domestic sewage was the main driver of downstream pollution.

[0187] 3. Analysis of spatial distribution characteristics:

[0188] SOM was used to analyze the spatial distribution characteristics of river water quality, and it was found that the COD concentration in the upper reaches was significantly higher than that in the middle and lower reaches, while ammonia nitrogen was concentrated in the middle and lower reaches.

[0189] 4. Spatial autocorrelation analysis:

[0190] The local Moran index is used to identify that the industrial areas upstream of the river are high pollution points (hot spots), while the living areas downstream become diffusion areas of pollution due to insufficient self-purification capacity.

[0191] Application effect:

[0192] Accurately identify pollution hotspots: clarify the scope of impact of industrial wastewater and domestic sewage on water quality.

[0193] Zoning management: Industrial enterprises in the upstream area are required to upgrade sewage treatment facilities and add sewage dams; downstream living areas implement sewage interception and management and ecological bank protection projects.

[0194] Improved water quality: After treatment, COD in the upper reaches decreased by 30%, and ammonia nitrogen concentration in the middle and lower reaches decreased by 25%. The overall water quality of the river improved, and some river sections reached Class III surface water standards.

[0195] The above embodiments demonstrate the successful application of this technical solution in the management of lakes and urban rivers. By accurately identifying pollution factors, analyzing the spatial distribution characteristics of water quality and quantifying spatial correlations, differentiated management measures are formulated, which significantly improves the management efficiency and effectiveness, and provides a scientific basis and technical support for water environment management.

[0196] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0197] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A quantitative analysis method for the spatial heterogeneity of water quality in a water area, characterized in that: include: S1, collect water samples from water areas regularly and analyze and test water quality indicators; S2, using a comprehensive assessment method to assess water quality and eutrophication status in water bodies; S3, factor identification analysis of water quality; S4, using self-organizing neural network to analyze the spatial distribution characteristics of water quality; S5, analyze whether there is spatial correlation in water quality and the degree of correlation through spatial autocorrelation.

2. The method for quantitatively analyzing the spatial heterogeneity of water quality in a water area according to claim 1, characterized in that: Comprehensive assessment methods include: water quality index and comprehensive trophic status index; The water quality index is used to comprehensively evaluate water quality, and its calculation formula is as follows: Where WQI is the comprehensive water quality index, Ci is the standardized score of water quality factor i, Pi is the weight of water quality factor i, the minimum value of Pi is 1, and the maximum value is 4; the WQI evaluation value range is 0-100, and the larger the value, the higher the water quality health; according to the score, water quality is divided into 5 levels: excellent [90-100], good [70-90], general [50-70], poor [25-50], and very poor [0-25).

3. The method for quantitatively analyzing the spatial heterogeneity of water quality in a water area according to claim 1, characterized in that: The comprehensive trophic state index is used to evaluate the eutrophication status of water bodies: the evaluation factors include ρ(Chla), ρ(TP), ρ(TN), SD, and ρ(CODMn); ​​taking ρ(Chla) as the benchmark parameter, the normalized weight calculation formula of the jth parameter is: Where rij is the correlation coefficient between the jth parameter and the benchmark parameter p(Chla), and m is the number of evaluation parameters; The calculation formulas of nutritional status index of each parameter are as follows: TLI(Chla)=10(2.5+1.086×ln[Chla]); TLI(TP)=10(9.436+1.624×ln[TP]); TLI(TN)=10(5.453+1.694×ln[TN]); TLI(SD)=10(5.118-1.94×ln[SD]); TLI(CODMn)=10(0.109+2.661×ln[CODMn]); Where: TLI(Chla), TLI(TP), TLI(TN), TLI(SD), TLI(CODMn) are the comprehensive nutrient status indexes of Chla, TP, TN, SD, and CODMn, respectively; [Chla] is ρ(Chla), mg / m3; [TP], [TN], [CODMn] are ρ(TP), ρ(TN), ρ(CODMn), mg / L, respectively; [SD] is SD, m; According to the score, the eutrophication status is divided into 5 levels: oligotrophic (0-30], mesotrophic [30-50], slightly eutrophic (50-60], moderately eutrophic (60-70], and severely eutrophic (70-100]; Conduct factor identification analysis on water quality, including: The three common factors of nitrogen and phosphorus indexes, physical indexes, and organic matter and eutrophication indexes were extracted to calculate the variance contribution rate.

4. The method for quantitatively analyzing the spatial heterogeneity of water quality in a water area according to claim 1, characterized in that: The specific analysis of water quality spatial distribution characteristics using self-organizing neural networks is as follows: By inputting the tested water quality sample data, the neural network will self-organize and automatically classify the training samples. The specific implementation process includes: Initialization of SOM model parameters: Initially assign a small random weight vector W = (Wi) to each neural unit, i = 1, 2, ..., m, where Wi = (ω1, ω2, ..., ωp); determine the initial value Ng(0) of the neighborhood Ng(t), and the learning rate is η; input the water quality sample feature data set X = (Xj), j = 1, 2, ...n, where Xj = (x1, x2, ..., xq); ② Vector normalization: Normalize the weight vector and sample vector to obtain Wi′=(ω1′,ω2′,…,ωp′) and Xj′=(x1′,x2′,…,xq′) respectively; ③ Find the winning neuron g: Find the minimum distance between the input sample vector Xj′ and the weight vector Wi′ corresponding to each neuron in the competition layer, and calculate the minimum Euclidean distance d between the two; dg = min[dj], j = 1, 2, ... m; ④ Adjust network output and weights: update weight Wj(t+1)=Wj(t)+α[Xj-Wj(t)], neighborhood Ng(t)=int{Ng(0)}[1-t / T]; ⑤ Repeat steps 3 and 4 until the number of learning times reaches T; Spatial autocorrelation is used to analyze whether there is spatial correlation and the degree of correlation in water quality. The degree of correlation is reflected by global Moran's I, and the calculation formula is as follows: Where: n is the number of spatial regions studied; Xi and Xj are the water quality of regions i and j respectively, and is the mean water quality of the entire study area; Wij is the spatial weight matrix; Moran's I is between -1 and 1. The closer the absolute value is to 1, the stronger the spatial autocorrelation is. A positive value indicates that Xi and Xj change in the same direction and the data are positively correlated. A negative value indicates that Xi and Xj change in opposite directions and the data are negatively correlated. A value of 0 indicates that the data is randomly distributed and has no autocorrelation.

5. The method for quantitatively analyzing the spatial heterogeneity of water quality in a water area according to claim 1, characterized in that: The data is searched for missing values, error values ​​are eliminated, and standardized to remove dimensions. Specifically: Before data analysis, SPSS19 software was used to find missing values, eliminate error values, and standardize the data to remove dimensions. The KS normality test of water quality data was performed at a significance level of 0.

05. Based on SPSS19, the spatial characteristics and main pollution factors of lake water quality were analyzed by factor analysis, and the KMO and spherical Bartlett test methods were used.

6. The method for quantitatively analyzing the spatial heterogeneity of water quality in a water area according to claim 1, characterized in that: Collect water samples regularly and analyze and test water quality indicators, specifically: Nine typical water quality indicators, including pH, ρ(DO), ρ(TP), ρ(TN), ρ(NH3-N), ρ(CODMn), ρ(SS), ρ(NO3-N), and ρ(Chla), were selected for analysis. Among them, pH and ρ(DO) were measured on-site using a portable dissolved oxygen meter, and other indicators were measured by refrigerating water samples and then bringing them back to the laboratory.

7. A system for quantitatively analyzing spatial differentiation characteristics of water quality in a water area, which implements the method for quantitatively analyzing spatial differentiation characteristics of water quality in a water area as claimed in any one of claims 1 to 6, characterized in that: include: Analysis and testing module, regularly collects water samples from water areas and analyzes and tests water quality indicators; Comprehensive assessment module, which uses comprehensive assessment methods to assess water quality and eutrophication status of water bodies; Factor identification and analysis module, which performs factor identification and analysis on water quality; Spatial distribution characteristics analysis module, using self-organizing neural network to analyze the spatial distribution characteristics of water quality; The correlation analysis module uses spatial autocorrelation to analyze whether there is spatial correlation in water quality and the degree of correlation.

8. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for quantitatively analyzing the spatial differentiation characteristics of water quality in a water area as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to execute the steps of the method for quantitatively analyzing the spatial heterogeneity characteristics of water quality in a water area as described in any one of claims 1 to 6.

10. An information data processing terminal, comprising the quantitative analysis system for spatial differentiation characteristics of water quality in a water area as claimed in claim 7.

Images