A method and system for analyzing the mechanism of water quality difference between river reaches considering river morphology characteristics
By constructing a water quality difference mechanism analysis method that takes into account river morphology characteristics, the problem that river morphology characteristics are not fully considered in the existing technology is solved, the accuracy and applicability of water quality difference analysis of river sections are improved, and the ability to reflect the synergistic change characteristics of multiple indicators is enhanced.
Patent Information
- Application Number
- CN202610264149.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-05
- Publication Date
- 2026-06-16
Smart Images
Figure CN122220747A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of water quality difference mechanism analysis technology, and in particular to a method and system for analyzing the mechanism of water quality differences between river sections considering river morphology characteristics. Background Technology
[0002] River water quality is a crucial evaluation indicator in water environment management and water resource protection, playing a vital role in ensuring regional ecological security, drinking water safety, and supporting comprehensive watershed management. In national or regional water environment management practices, different river sections exhibit significant differences in water quality status and variation characteristics due to variations in natural conditions and the intensity of human activities. Systematic analysis of water quality differences among multiple river sections helps reveal the varying responses of pollutants in different river sections, providing technical support for regional watershed management and refined water environment governance.
[0003] Existing methods for analyzing the mechanisms of water quality differences between river sections typically rely on water quality monitoring data, combined with meteorological factors, hydrological factors, and information on pollution source emissions. They employ statistical analysis models or water quality models to analyze the mechanisms of these differences. In regional or national-scale studies, some methods also incorporate indicators such as land use type and human activity intensity to reflect the impact of human activities on water quality between river sections. These methods have been widely used in comparing and analyzing the changing trends of water quality across multiple river sections.
[0004] However, current technologies for analyzing the mechanisms of water quality differences between river sections primarily focus on the influence of external driving factors, with relatively insufficient consideration given to the river's own physical characteristics. River morphology features such as channel width and depth directly affect the cross-sectional area of the water body, flow conditions, and the dilution and transport processes of pollutants, significantly influencing the mechanisms of water quality differences between river sections. However, in existing analytical methods, these are often treated merely as background information or auxiliary descriptive factors, rather than being systematically incorporated into the water quality change analysis process as independent influencing factors. In multi-section, cross-regional application scenarios, this can easily lead to insufficient explanatory power for the mechanisms of water quality differences between different river sections, affecting the accuracy and reliability of the analytical results.
[0005] Furthermore, existing water quality analysis methods still have certain limitations in handling multiple indicators. One type of method typically calculates multiple water quality indicators into a single water quality evaluation index for analysis. While this facilitates overall evaluation, it can easily obscure the differences and interactions between different water quality indicators. Another type of method establishes independent models for each water quality indicator, lacking a holistic characterization of the changing relationships among multiple indicators. In the analysis of river water quality changes at national or regional scales, the above methods are insufficient to fully reflect the coordinated changes of multiple water quality indicators between upstream and downstream sections of different river segments. Summary of the Invention
[0006] To overcome the shortcomings of the above-mentioned analysis of river water quality that does not consider river morphology, this invention provides a method and system for analyzing the mechanism of water quality differences between river sections that takes into account river morphology.
[0007] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: This invention provides a method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics, including: Obtain water quality parameters from upstream and downstream of the target river section and construct a multi-index output vector; Obtain water quality influencing factor data for the target river section and construct a feature input vector. The water quality influencing factor data includes river morphology feature factors. The multi-index output vector and feature input vector are preprocessed to obtain the preprocessed multi-index output vector and feature input vector; Based on the preprocessed multi-index output vector and feature input vector, a water quality difference mechanism analysis model is constructed. Based on the water quality difference mechanism analysis model, the analysis results of the water quality difference mechanism in the target river section are output.
[0008] Preferably, water quality parameters of the upstream and downstream sections of the target river are obtained, and a multi-index output vector is constructed, including: Obtain the k-th water quality parameter of the upstream and downstream sections of the target river segment s, and set them as follows: and Where t is the time index; The relative change rate of the kth water quality indicator is:
[0009] Constructing multiple output vectors based on the relative rate of change of water quality:
[0010] in, This is a multi-index output vector.
[0011] Preferably, the water quality influencing factor data also includes meteorological factors, hydrological factors, and human activity factors.
[0012] Preferably, based on the water quality influencing factor data, a feature input vector is constructed, including:
[0013] in, For the feature input vector, Meteorological factors, Hydrological factors, As a factor of human activity, For river channel morphology characteristics; Among them, river channel morphology characteristic factors The expression is:
[0014] in, The width of the river channel, For river depth, .
[0015] Preferably, the multi-index output vector and feature input vector are preprocessed to obtain preprocessed multi-index output vector and feature input vector, including: Missing values are removed from the multi-index output vector and feature input vector to obtain a missing multi-index output vector and feature input vector; The feature input vector with missing values removed is then subjected to outlier removal to obtain the feature input vector with outlier removal. Robust scaling is applied to the feature input vector for outlier removal, and standardization is applied to the multi-index output vector for missing value removal.
[0016] in, For robust scaling operations, For standardized operation, For the preprocessed feature input vector, This is the preprocessed multi-index output vector.
[0017] Preferably, based on the preprocessed multi-index output vector and feature input vector, a water quality difference mechanism analysis model is constructed, including:
[0018] in, A model for analyzing the mechanism of water quality differences. Let I be the noise variance, and I be the identity matrix.
[0019] Preferably, based on the water quality difference mechanism analysis model, the water quality difference mechanism analysis results of the target river section are output, including: The hyperparameters to be solved in the water quality difference mechanism analysis model are determined by using the multi-output covariance function and the automatic correlation to determine the squared exponent kernel function. Based on the preprocessed multi-index output vector and feature input vector, the hyperparameters to be solved are obtained, and the water quality difference mechanism analysis results of the target river section are output according to the solved hyperparameters.
[0020] Preferably, the hyperparameters to be solved in the water quality difference mechanism analysis model are determined using a multi-output covariance function and an automatic correlation-determined squared exponent kernel function, including: The expression for the multi-output covariance function is as follows:
[0021] in, For a collaborative regionalized matrix, For Kronecker product; The expression for the autocorrelation-determining squared exponential kernel function is as follows:
[0022] in, Let be the signal variance, and p be the dimension of the influencing factor. and These are the feature input vectors for outlier removal and the preprocessed feature input vectors, respectively. Let be the length scale parameter of the d-th influence factor; The cooperative regionalization matrix Length scale parameters It has been identified as a hyperparameter to be solved.
[0023] Preferably, based on the preprocessed multi-index output vector and feature input vector, the hyperparameters to be solved are obtained, and the water quality difference mechanism analysis results of the target river section are output according to the solved hyperparameters, including: The water quality difference mechanism analysis model is solved by maximizing the marginal likelihood function to obtain the solved co-regionalized matrix and length scale parameters. Based on the solved co-regionalized matrix, the correlation coefficients between the variations of different water quality indicators are calculated. Based on the solved length scale parameters, calculate the weight sensitivity of each influencing factor; Based on the correlation coefficients between variations of different water quality indicators and the weight sensitivity of each influencing factor, the analysis results of the water quality difference mechanism in the target river section are output.
[0024] This invention also provides a system for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics, comprising: The multi-index output vector construction module is used to obtain water quality parameters of the upstream and downstream of the target river section and construct multi-index output vectors. The feature input vector construction module is used to obtain water quality influencing factor data of the target river section and construct feature input vectors. The water quality influencing factor data includes river morphology feature factors. The preprocessing module is used to preprocess the multi-index output vector and feature input vector to obtain the preprocessed multi-index output vector and feature input vector. The water quality difference mechanism analysis model construction module is used to construct a water quality difference mechanism analysis model based on the preprocessed multi-index output vector and feature input vector. The water quality difference mechanism analysis result module is used to output the water quality difference mechanism analysis results of the target river section based on the water quality difference mechanism analysis model.
[0025] Compared with the prior art, the beneficial effects of the technical solution of the present invention are: This invention provides a method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics. First, upstream and downstream paired units of the target river segment are constructed. Then, based on these paired units, multi-index output vectors and feature input vectors are constructed. Next, the multi-index output vectors and feature input vectors are preprocessed to obtain preprocessed multi-index output vectors and feature input vectors. Based on these preprocessed multi-index output vectors and feature input vectors, a water quality difference mechanism analysis model is constructed. Finally, based on the water quality difference mechanism analysis model, the analysis results of the water quality difference mechanism of the target river segment are output. This invention improves the accuracy and applicability of analyzing the mechanism of water quality differences between river segments by introducing river morphology characteristics as an independent influencing factor into the water quality difference analysis process. Attached Figure Description
[0026] Figure 1 This is a flowchart illustrating a method for analyzing the mechanism of water quality differences between river segments that considers river morphology characteristics, as shown in Example 1. Figure 2 This is a schematic diagram of the structure of a system for analyzing the mechanism of water quality differences between river sections that takes into account the morphological characteristics of the river channel, as shown in Example 3. Detailed Implementation
[0027] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent. To better illustrate this embodiment, some parts in the accompanying drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions; It will be understood by those skilled in the art that certain well-known structures and their descriptions may be omitted in the accompanying drawings.
[0028] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0029] Example 1 This embodiment provides a method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics, such as... Figure 1 As shown, it includes: Obtain water quality parameters from upstream and downstream of the target river section and construct a multi-index output vector; Obtain water quality influencing factor data for the target river section and construct a feature input vector. The water quality influencing factor data includes river morphology feature factors. The multi-index output vector and feature input vector are preprocessed to obtain the preprocessed multi-index output vector and feature input vector; Based on the preprocessed multi-index output vector and feature input vector, a water quality difference mechanism analysis model is constructed. Based on the water quality difference mechanism analysis model, the analysis results of the water quality difference mechanism in the target river section are output.
[0030] In its implementation, this application first obtains water quality parameters from the upstream and downstream of the target river section and constructs a multi-index output vector. Next, it obtains water quality influencing factor data for the target river section and constructs a feature input vector, including river channel morphology characteristic factors. Then, it preprocesses the multi-index output vector and feature input vector to obtain preprocessed multi-index output vector and feature input vector. Based on the preprocessed multi-index output vector and feature input vector, it constructs a water quality difference mechanism analysis model. Finally, based on the water quality difference mechanism analysis model, it outputs the water quality difference mechanism analysis results for the target river section. This invention improves the accuracy and applicability of river section water quality difference mechanism analysis by introducing river channel morphology characteristics and incorporating them as independent influencing factors into the water quality difference analysis process.
[0031] Example 2 This embodiment provides a method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics, including: Obtain water quality parameters from upstream and downstream of the target river section and construct a multi-index output vector; Obtain water quality influencing factor data for the target river section and construct a feature input vector. The water quality influencing factor data includes river morphology feature factors. The multi-index output vector and feature input vector are preprocessed to obtain the preprocessed multi-index output vector and feature input vector; Based on the preprocessed multi-index output vector and feature input vector, a water quality difference mechanism analysis model is constructed. Based on the water quality difference mechanism analysis model, the analysis results of the water quality difference mechanism in the target river section are output.
[0032] It should be noted that in this embodiment, water quality parameters of the upstream and downstream of the target river section are obtained, and a multi-index output vector is constructed, including: Obtain the k-th water quality parameter of the upstream and downstream sections of the target river segment s, and set them as follows: and Where t is the time index; The relative change rate of the kth water quality indicator is:
[0033] Constructing multiple output vectors based on the relative rate of change of water quality:
[0034] in, This is a multi-index output vector.
[0035] It should be noted that, in this embodiment, the water quality influencing factor data also includes meteorological factors, hydrological factors, and human activity factors.
[0036] It should be noted that, in this embodiment, the feature input vector is constructed based on the water quality influencing factor data, including:
[0037] in, For the feature input vector, Meteorological factors, Hydrological factors, As a factor of human activity, For river channel morphology characteristics; Among them, river channel morphology characteristic factors The expression is:
[0038] in, The width of the river channel, For river depth, .
[0039] It should be noted that, in this embodiment, the multi-index output vector and feature input vector are preprocessed to obtain preprocessed multi-index output vector and feature input vector, including: Missing values are removed from the multi-index output vector and feature input vector to obtain a missing multi-index output vector and feature input vector; The feature input vector with missing values removed is then subjected to outlier removal to obtain the feature input vector with outlier removal. Robust scaling is applied to the feature input vector for outlier removal, and standardization is applied to the multi-index output vector for missing value removal.
[0040] in, For robust scaling operations, For standardized operation, For the preprocessed feature input vector, This is the preprocessed multi-index output vector.
[0041] It should be noted that, in this embodiment, a water quality difference mechanism analysis model is constructed based on the preprocessed multi-index output vector and feature input vector, including:
[0042] in, A model for analyzing the mechanism of water quality differences. Let I be the noise variance, and I be the identity matrix.
[0043] It should be noted that, in this embodiment, based on the water quality difference mechanism analysis model, the output of the water quality difference mechanism analysis results for the target river section includes: The hyperparameters to be solved in the water quality difference mechanism analysis model are determined by using the multi-output covariance function and the automatic correlation to determine the squared exponent kernel function. Based on the preprocessed multi-index output vector and feature input vector, the hyperparameters to be solved are obtained, and the water quality difference mechanism analysis results of the target river section are output according to the solved hyperparameters.
[0044] It should be noted that, in this embodiment, the multi-output covariance function and the automatic correlation-determining squared exponent kernel function are used to determine the hyperparameters to be solved in the water quality difference mechanism analysis model, including: The expression for the multi-output covariance function is as follows:
[0045] in, For a collaborative regionalized matrix, For Kronecker product; The expression for the autocorrelation-determining squared exponential kernel function is as follows:
[0046] in, Let be the signal variance, and p be the dimension of the influencing factor. and These are the feature input vectors for outlier removal and the preprocessed feature input vectors, respectively. Let be the length scale parameter of the d-th influence factor; The cooperative regionalization matrix Length scale parameters It has been identified as a hyperparameter to be solved.
[0047] It should be noted that, in this embodiment, based on the preprocessed multi-index output vector and feature input vector, the hyperparameters to be solved are obtained, and the water quality difference mechanism analysis results of the target river section are output according to the solved hyperparameters, including: The water quality difference mechanism analysis model is solved by maximizing the marginal likelihood function to obtain the solved co-regionalized matrix and length scale parameters. Based on the solved co-regionalized matrix, the correlation coefficients between the variations of different water quality indicators are calculated. Based on the solved length scale parameters, calculate the weight sensitivity of each influencing factor; Based on the correlation coefficients between variations of different water quality indicators and the weight sensitivity of each influencing factor, the analysis results of the water quality difference mechanism in the target river section are output.
[0048] Example 3 This embodiment provides a system for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics, used to implement the method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics described in Embodiment 1 or 2. Figure 2 As shown, it includes: The multi-index output vector construction module is used to obtain water quality parameters of the upstream and downstream of the target river section and construct multi-index output vectors. The feature input vector construction module is used to obtain water quality influencing factor data of the target river section and construct feature input vectors. The water quality influencing factor data includes river morphology feature factors. The preprocessing module is used to preprocess the multi-index output vector and feature input vector to obtain the preprocessed multi-index output vector and feature input vector. The water quality difference mechanism analysis model construction module is used to construct a water quality difference mechanism analysis model based on the preprocessed multi-index output vector and feature input vector. The water quality difference mechanism analysis result module is used to output the water quality difference mechanism analysis results of the target river section based on the water quality difference mechanism analysis model.
[0049] The same or similar labels correspond to the same or similar parts; The terms used to describe positional relationships in the accompanying drawings are for illustrative purposes only and should not be construed as limiting this patent. Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.
Claims
1. A method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics, characterized in that, include: Obtain water quality parameters from upstream and downstream of the target river section and construct a multi-index output vector; Obtain water quality influencing factor data for the target river section and construct a feature input vector. The water quality influencing factor data includes river morphology feature factors. The multi-index output vector and feature input vector are preprocessed to obtain the preprocessed multi-index output vector and feature input vector; Based on the preprocessed multi-index output vector and feature input vector, a water quality difference mechanism analysis model is constructed. Based on the water quality difference mechanism analysis model, the analysis results of the water quality difference mechanism in the target river section are output.
2. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 1, characterized in that, Obtain water quality parameters from upstream and downstream of the target river section, and construct a multi-index output vector, including: Obtain the k-th water quality parameter of the upstream and downstream sections of the target river segment s, and set them as follows: and Where t is the time index; The relative change rate of the kth water quality indicator is: Constructing multiple output vectors based on the relative rate of change of water quality: in, This is a multi-index output vector.
3. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 2, characterized in that, The water quality influencing factor data also include meteorological factors, hydrological factors, and human activity factors.
4. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 3, characterized in that, Based on the water quality influencing factor data, a feature input vector is constructed, including: in, For the feature input vector, Meteorological factors, Hydrological factors, As a factor of human activity, For river channel morphology characteristics; Among them, river channel morphology characteristic factors The expression is: in, The width of the river channel, For river depth, .
5. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 4, characterized in that, The multi-index output vector and feature input vector are preprocessed to obtain preprocessed multi-index output vector and feature input vector, including: Missing values are removed from the multi-index output vector and feature input vector to obtain a missing multi-index output vector and feature input vector; The feature input vector with missing values removed is then subjected to outlier removal to obtain the feature input vector with outlier removal. Robust scaling is applied to the feature input vector for outlier removal, and standardization is applied to the multi-index output vector for missing value removal. in, For robust scaling operations, For standardized operation, For the preprocessed feature input vector, This is the preprocessed multi-index output vector.
6. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 5, characterized in that, Based on the preprocessed multi-index output vector and feature input vector, a water quality difference mechanism analysis model is constructed, including: in, A model for analyzing the mechanism of water quality differences. Let I be the noise variance, and I be the identity matrix.
7. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 6, characterized in that, Based on the aforementioned water quality difference mechanism analysis model, the analysis results of the water quality difference mechanism in the target river section are output, including: The hyperparameters to be solved in the water quality difference mechanism analysis model are determined by using the multi-output covariance function and the automatic correlation to determine the squared exponent kernel function. Based on the preprocessed multi-index output vector and feature input vector, the hyperparameters to be solved are obtained, and the water quality difference mechanism analysis results of the target river section are output according to the solved hyperparameters.
8. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 7, characterized in that, Using the multi-output covariance function and the autocorrelation-determined squared exponent kernel function, the hyperparameters to be solved in the water quality difference mechanism analysis model are determined, including: The expression for the multi-output covariance function is as follows: in, For a collaborative regionalized matrix, For Kronecker product; The expression for the autocorrelation-determining squared exponential kernel function is as follows: in, Let be the signal variance, and p be the dimension of the influencing factor. and These are the feature input vectors for outlier removal and the preprocessed feature input vectors, respectively. Let be the length scale parameter of the d-th influence factor; The cooperative regionalization matrix Length scale parameters It has been identified as a hyperparameter to be solved.
9. The method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics according to claim 8, characterized in that, Based on the preprocessed multi-index output vector and feature input vector, the hyperparameters to be solved are obtained, and the analysis results of the water quality difference mechanism of the target river section are output according to the solved hyperparameters, including: The water quality difference mechanism analysis model is solved by maximizing the marginal likelihood function to obtain the solved co-regionalized matrix and length scale parameters. Based on the solved co-regionalized matrix, the correlation coefficients between the variations of different water quality indicators are calculated. Based on the solved length scale parameters, calculate the weight sensitivity of each influencing factor; Based on the correlation coefficients between variations of different water quality indicators and the weight sensitivity of each influencing factor, the analysis results of the water quality difference mechanism in the target river section are output.
10. A system for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics, used to implement the method for analyzing the mechanism of water quality differences between river segments considering river morphology characteristics as described in claims 1-9, characterized in that, include: The multi-index output vector construction module is used to obtain water quality parameters of the upstream and downstream of the target river section and construct multi-index output vectors. The feature input vector construction module is used to obtain water quality influencing factor data of the target river section and construct feature input vectors. The water quality influencing factor data includes river morphology feature factors. The preprocessing module is used to preprocess the multi-index output vector and feature input vector to obtain the preprocessed multi-index output vector and feature input vector. The water quality difference mechanism analysis model construction module is used to construct a water quality difference mechanism analysis model based on the preprocessed multi-index output vector and feature input vector. The water quality difference mechanism analysis result module is used to output the water quality difference mechanism analysis results of the target river section based on the water quality difference mechanism analysis model.