Drainage basin multi-region water resource competition relationship dynamic quantification and risk early warning method and system

Abstract

The invention discloses a watershed multi-region water resource competition relationship dynamic quantification and risk early warning method and system. The method comprises the following steps: determining regions and departments participating in water resource competition in a watershed; collecting data over the years of water supply amount, water consumption amount, population amount and economic development level of each region and environmental influence data of geographic position, climate change and renewable water resource amount; constructing a competition intensity matrix, and identifying key competition area pairs; constructing and adopting a parameter inversion method to solve a competition coefficient in the multi-region water resource competition differential dynamical model; dynamically quantifying the influence degree, direction and ecological relationship type of inter-regional water resource competition based on a set competition and cooperation intensity grade, and analyzing a water resource competition pattern between drainage basin regions; and setting different scenes in the future, inputting driving parameters into the calibrated multi-region water resource competition differential dynamic model, simulating and predicting the development situation of the water resource competition relationship in different scenes in the future, and carrying out risk monitoring.

Description

Technical Field

[0001] This invention belongs to the field of water resources management technology, specifically relating to a method and system for dynamic quantification and risk early warning of water resources competition relationships in multiple regions of a river basin. Background Technology

[0002] With global climate change, accelerated socio-economic development, and continuous population growth, the contradiction between the uneven spatial and temporal distribution of water resources, a fundamental strategic resource, and the rigid increase in demand is becoming increasingly acute. At the watershed scale, competition for limited water resources among different administrative regions upstream and downstream is intensifying due to differences in resource endowment and development stages, forming a complex cross-regional water resource competition pattern. If such competition is not managed properly, it will not only restrict the coordinated development of regional economies but may even trigger social conflicts, threatening watershed water security and ecological security.

[0003] Currently, quantitative analysis methods in water resource management are mostly focused on supply and demand balance analysis in single regions, rule-based optimization allocation, and reservoir scheduling simulation. While these traditional methods have played an important role in water resource planning, their inherent limitations are becoming increasingly apparent. First, traditional methods are mostly static or quasi-static analysis paradigms, making it difficult to characterize the dynamic interaction and feedback mechanisms that arise when multiple regions act as competitors in water use. Since water resource competition is a complex and dynamically evolving process, static snapshots cannot capture its inherent laws. Second, traditional methods emphasize the overall supply and demand balance of the system or treat regions as isolated entities, lacking the ability to characterize the two-way, quantitative relationship of "how water use behavior in region A specifically affects region B," and failing to answer core questions such as "where are the key points of competition" and "how intense is the competition." Finally, due to the lack of description of competition dynamics in the models themselves, existing technologies are unable to support the scientific prediction of water resource competition under different future development scenarios (such as extreme climates or rapid economic growth), and are even less able to achieve forward-looking risk warnings based on quantitative indicators, causing management decisions to often fall into a reactive mode.

[0004] Population dynamics in ecology provides a mature mathematical framework for describing competitive behavior in environments with limited resources. This theory can subtly characterize the dynamic relationships—competition, symbiosis, or neutrality—that arise between different species vying for the same resource. In recent years, some scholars have attempted to apply the ideas of this theory to resource management fields such as energy and fisheries, offering a promising new perspective for quantifying multi-regional water resource competition. However, successfully applying population dynamics to water resource management still faces a core bottleneck: how to accurately solve for key parameters in the model (such as competition coefficients) to accurately quantify inter-regional competition. Existing parameter determination methods generally suffer from insufficient accuracy, poor universality, or excessively high data requirements, resulting in low reliability of model outputs, making them difficult to implement in practice, and lacking technical design for linkage with risk warning systems.

[0005] Therefore, it is particularly urgent to develop an innovative method that can overcome the above-mentioned defects. This method can not only inherit the mechanistic advantages of ecological population dynamics models in describing competitive relationships, but also break through the technical bottleneck of parameter solving, and ultimately achieve accurate dynamic quantification and scientific risk warning of water resource competition relationships in multiple regions. Summary of the Invention

[0006] To overcome the shortcomings of existing technologies, this invention provides a method for dynamic quantification and risk early warning of water resource competition relationships in multiple regions of a watershed. This method can accurately analyze the degree, direction, and pattern of water resource competition among regions, simulate and predict the evolution of water resource competition relationships under different future scenarios, and ultimately achieve scientific risk early warning based on quantitative indicators, providing a basis for decision-making in the proactive and precise management of water resources.

[0007] Another objective of this invention is to provide a system that employs a dynamic quantification and risk early warning method for multi-regional water resource competition relationships within a watershed.

[0008] The technical solution of the present invention to solve the above-mentioned technical problems is:

[0009] A method for dynamic quantification and risk early warning of water resource competition relationships in multiple regions of a watershed includes the following steps:

[0010] S1: Based on the administrative division, identify the areas within the basin that participate in water resource competition, and based on the departmental division, identify the agricultural, industrial, domestic, and ecological water use departments that participate in water resource competition within the area;

[0011] S2: Collect historical data on water supply, water consumption, population, and economic development level for each region, as well as data on geographical location, climate change, and environmental impact of renewable water resources, and preprocess the data.

[0012] S3: Calculate the water shortage rate of each region, and based on the calculated water shortage rate of each region, calculate the water resource competition intensity between any two regions to construct a competition intensity matrix. Analyze the constructed competition intensity matrix to identify key competitive region pairs.

[0013] S4: Based on the theory of ecological population competition dynamics, construct a differential dynamics model of water resource competition in multiple regions;

[0014] S5: The competition coefficients in the multi-region water resource competition differential dynamics model are solved using the parameter inversion method;

[0015] S6: Based on the water resource competition coefficient and the set competition and cooperation intensity levels, dynamically quantify the impact degree, direction and ecological relationship type of water resource competition between regions, and systematically analyze the water resource competition pattern between watershed regions.

[0016] S7. Set different future scenarios and input the driving parameters of different scenarios into the calibrated multi-regional water resource competition differential dynamics model to simulate and predict the development trend of water resource competition relationship under different future scenarios. When the predicted value of the competition coefficient is greater than or equal to the preset threshold, the corresponding level of risk warning is triggered.

[0017] Preferably, in step S2, the preprocessing includes data cleaning and processing of missing and outlier values.

[0018] Preferably, in step S3, the steps for obtaining the competition intensity matrix are as follows:

[0019] The water shortage rate of a region is obtained by calculating the ratio between the difference between the region's annual water demand and annual water supply and the region's annual water supply. The intensity of water competition between two adjacent regions is determined by finding the first absolute value of the difference between the water shortage rates of any two regions and the second absolute value of the sum of the water shortage rates of the two regions, and by finding the ratio between the first absolute value and the second absolute value.

[0020] Based on the calculated water competition intensity between any two regions, a competition intensity matrix is ​​constructed:

[0021] ;

[0022] In the formula: For each element in the competition intensity matrix, represents the intensity of the interaction between the i-th region and the j-th region. for The competition intensity matrix.

[0023] Preferably, in step S4, the multi-regional water resource competition differential dynamics model is constructed as follows:

[0024] ;

[0025] In the formula: Let be the water consumption of the i-th region at time t. Let be the annual growth rate of the i-th region at time t. Let be the competition coefficient between region j and region i at time t. Let n be the maximum allowable water consumption of the i-th region at time t, i.e., the environmental carrying capacity, where n is the total number of regions and t is the time in years.

[0026] Preferably, in step S5, the parameter inversion method is one or more of the following: grey system theory method, Bayesian inference method, genetic algorithm or machine learning regression algorithm.

[0027] Preferably, in step S6, a competition coefficient matrix is ​​constructed based on the solved competition coefficients; according to the magnitude and sign of each competition coefficient in the competition coefficient matrix, and referring to a preset competition and cooperation intensity grading standard table, the water resource relationship between any two regions is classified into specific qualitative levels, and simultaneously, according to the sign and numerical range of the competition coefficients, they are mapped to the defined ecological relationship types; by comprehensively analyzing the levels and relationship types of all region pairs, the competitive pattern of water resources across the entire region is systematically analyzed and clarified; wherein, the construction of the competition coefficient matrix... ;

[0028] ;

[0029] In the formula: The competition coefficient value in the competition coefficient matrix represents the degree of influence of the j-th region on the i-th region.

[0030] Preferably, in the competition and cooperation intensity grading standard table, the relationships are divided into competitive, cooperative, and no relationship based on the range of the competition coefficient; a competition coefficient in (-∞, 0) indicates a cooperative relationship; a competition coefficient in (0, +∞) indicates a competitive relationship; and a competition coefficient of 0 indicates that there is neither competition nor cooperation between the two regions.

[0031] In cooperative relationships, (-∞, -2] is set as deep cooperative, (-2, -1.5] as heavy cooperative, (-1.5, -1] as medium cooperative, and (-1, 0) as light cooperative; in competitive relationships, (0, 1) is set as light competitive, [1, 1.5) as medium competitive, [1.5, 2) as heavy competitive, and [2, +∞) as fierce competitive.

[0032] Preferably, the definition rules for ecological relationship types are as follows:

[0033] Using the competition coefficient value β between the i-th region and the j-th region, and the competition coefficient value α between the j-th region and the i-th region, a planar coordinate system is constructed.

[0034] When the planar coordinates are in the first quadrant, that is, both α and β are positive, the ecological relationship between the i-th region and the j-th region is a competitive relationship;

[0035] When the planar coordinates are in the second quadrant, that is, α is negative and β is positive, where β being positive means that the i-th region is in competition with the j-th region, and α being negative means that the j-th region is in cooperation with the i-th region. Therefore, it is inferred that the ecological relationship between the i-th region and the j-th region is a parasitic relationship, where the i-th region is the dominant party and the j-th region is the damaging party.

[0036] When the planar coordinates are in the third quadrant, that is, α is negative and β is negative, where β being negative means that the i-th region has a cooperative relationship with the j-th region, and α being negative means that the j-th region has a cooperative relationship with the i-th region. Therefore, it is inferred that the ecological relationship between the i-th region and the j-th region is a symbiotic relationship.

[0037] When the planar coordinates are in the fourth quadrant, i.e., α is positive and β is negative, where negative β indicates a cooperative relationship between region i and region j, and positive α indicates a competitive relationship between region j and region i. Therefore, it can be inferred that the ecological relationship between region i and region j is a parasitic relationship; where region i is the damaging party and region j is the dominant party.

[0038] When the planar coordinates are on the ordinate axis between the first and second quadrants, i.e. β is a positive number, the ecological relationship between the i-th region and the j-th region is a partial harm-symbiotic relationship.

[0039] When the planar coordinates lie on the ordinate axis between the third and fourth quadrants, β is negative. The ecological relationship between the i-th and j-th regions is a symbiotic relationship.

[0040] When the planar coordinates lie on the horizontal axis between the first and fourth quadrants, α is a positive number. The ecological relationship between the i-th and j-th regions is a symbiotic relationship of partial harm.

[0041] When the planar coordinates are on the horizontal axis between the second and third quadrants, i.e. α is negative, the ecological relationship between the i-th region and the j-th region is a symbiotic relationship.

[0042] When the planar coordinates are at the origin, that is, when both α and β are 0, the ecological relationship between the i-th region and the j-th region is a neutral relationship.

[0043] Preferably, in step S7, the designed future scenarios include a baseline scenario, a high-speed economic development scenario, a climate change drought scenario, and a strict water conservation policy scenario.

[0044] A dynamic quantification and risk early warning system for multi-regional water resource competition in a river basin includes:

[0045] Region and sector identification module: Based on administrative divisions, identify the regions within the basin that participate in water resource competition, and based on sectoral divisions, identify the agricultural, industrial, domestic, and ecological water use sectors that participate in water resource competition within the region;

[0046] Data Acquisition and Preprocessing Module: Collects historical data on water supply, water consumption, population, and economic development level for each region, as well as data on geographical location, climate change, and environmental impact of renewable water resources, and preprocesses the data.

[0047] Competition Intensity Matrix Construction and Key Region Pair Identification Module: Calculates the water shortage rate of each region, and calculates the water resource competition intensity between any two regions based on the calculated water shortage rate of each region, thereby constructing a competition intensity matrix. The constructed competition intensity matrix is ​​then analyzed to identify key competitive region pairs.

[0048] Differential dynamics model construction module: Based on the theory of ecological population competition dynamics, construct a multi-regional differential dynamics model of water resource competition;

[0049] Competition coefficient solution module: The competition coefficient in the multi-region water resource competition differential dynamics model is solved using the parameter inversion method;

[0050] Competition Pattern Analysis Module: Based on the competition coefficient of water resources and the set competition and cooperation intensity levels, the module dynamically quantifies the degree of impact, direction and ecological relationship type of competition between regions, and systematically analyzes the water resource competition pattern between watershed regions.

[0051] Scenario Simulation and Risk Warning Module: Set different future scenarios, input the driving parameters of different scenarios into the calibrated model, simulate and predict the development trend of water resource competition under different future scenarios, and trigger the corresponding level of risk warning when the predicted value of the competition coefficient reaches or exceeds the preset threshold.

[0052] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0053] 1. This invention breaks through the limitations of traditional static analysis, realizes dynamic quantification and forward-looking early warning of water resource competition, and provides core technical support for the scientific allocation of water resources and the early identification and collaborative management of cross-regional water resource competition. It is especially suitable for precise regulation and risk prevention in water-scarce areas.

[0054] 2. This invention achieves a deep integration of ecological theory and social water cycle issues by systematically applying the competition model in ecological population dynamics to the quantitative analysis of water resource competition relationships in multiple regions of a watershed.

[0055] 3. This invention proposes a competitive coefficient solution framework with parameter inversion as its core, and clarifies multiple applicable paths such as grey system theory, Bayesian inference, genetic algorithm and machine learning, which can effectively solve the industry problem of difficult accurate identification of key parameters in mechanism models.

[0056] 4. This invention constructs a closed-loop technical system from dynamic quantification to scenario early warning. By combining the competition coefficient with the preset competition / cooperation intensity grading standard, it realizes semantic analysis and level determination of the competitive landscape, and seamlessly links with the risk early warning mechanism.

[0057] 5. This invention upgrades water resource competition analysis from traditional static snapshots to continuous dynamic simulation, which can reveal the evolution process and inherent laws of competitive relationships; through mathematical modeling and parameter inversion, it can accurately quantify the intensity and direction of competition between regions, transforming qualitative descriptions of competition into quantitative decision indicators.

[0058] 6. This invention has the ability to predict the competitive landscape under different future development paths and the function of early risk warning, enabling water resource management to shift from passive response to proactive intervention.

[0059] 7. This invention provides managers with multi-dimensional and visualized decision-making information such as "key conflict area pairs", "competition level map" and "risk warning signals", which greatly improves the scientificity and efficiency of water resource allocation and conflict mediation. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating the method for dynamic quantification and risk early warning of water resource competition relationships in multiple regions of a river basin, as presented in this invention.

[0061] Figure 2 This diagram illustrates the method for dynamic quantification and risk early warning of water resource competition relationships in multiple regions of a watershed, as presented in this invention. Detailed Implementation

[0062] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0063] Example 1

[0064] See Figures 1-2 The present invention provides a method for dynamic quantification and risk early warning of multi-regional water resource competition in a watershed, comprising the following steps:

[0065] S1. Define the areas of competition for water use:

[0066] Based on the administrative divisions, the areas within the basin that participate in water resource competition are identified, and based on the departmental divisions, the agricultural, industrial, domestic, and ecological water use departments that participate in water resource competition within the area are identified;

[0067] S2. Acquisition and preprocessing of multi-source data:

[0068] Collect historical data on water supply, water consumption, population, and economic development level in various regions, as well as data on geographical location, climate change, and environmental impact of renewable water resources, and preprocess the data.

[0069] In this embodiment, the preprocessing includes data cleaning and handling missing and outlier values.

[0070] S3. Identify key competitive regions:

[0071] Using preprocessed historical water supply and demand data for each region, the water shortage rate for each region is calculated. Based on the calculated water shortage rate, the water resource competition intensity between any two regions is calculated to construct a competition intensity matrix. The constructed competition intensity matrix is ​​then analyzed to identify key competitive region pairs.

[0072] Water shortage rate in a single area The calculation formula is:

[0073] ;

[0074] In the formula: Let i be the water shortage rate of region i. Let be the annual water demand of region i. Let i be the annual water supply for region i;

[0075] Water competition intensity between any two regions The calculation formula is:

[0076] ;

[0077] In the formula: The intensity of water competition between the two regions, and The value range is [0,1];

[0078] Based on the calculated water competition intensity between any two regions, a competition intensity matrix is ​​constructed:

[0079] ;

[0080] In the formula: For each element in the competition intensity matrix, represents the intensity of the interaction between the i-th region and the j-th region. for The competition intensity matrix.

[0081] S4. Construct a multi-regional water resource competition model:

[0082] Based on the theory of ecological population competition dynamics, a differential dynamics model of water resource competition in multiple regions is constructed, as follows:

[0083] Based on the theory of ecological population competition dynamics, the variables and parameters in the model are determined, including the annual water consumption, annual water consumption growth rate, and environmental carrying capacity of each region, and a multi-regional water resource competition differential dynamics model is constructed: The constructed multi-regional water resource competition differential dynamics model is as follows:

[0084] ;

[0085] In the formula: Let be the water consumption of the i-th region at time t. Let be the annual growth rate of the i-th region at time t. Let be the competition coefficient between region j and region i at time t. Let n be the maximum allowable water consumption of the i-th region at time t, i.e., the environmental carrying capacity, where n is the total number of regions and t is the time in years.

[0086] S5. Inversion solution for competition coefficients:

[0087] The competition coefficients in the multi-regional water resource competition differential dynamics model are solved using a parameter inversion method. The parameter inversion method can be selected from one or more of the following: grey system theory method, Bayesian inference method, genetic algorithm, or machine learning regression algorithm.

[0088] When the parameter inversion method is selected as the grey system theory method, it specifically includes:

[0089] (1) The multi-regional water resource competition differential dynamics model is decoupled into a subset of competition relationships between two regions under specific conditions;

[0090] (2) The historical water consumption data sequence of each competitive subset is accumulated to reduce randomness;

[0091] (3) Establish the GM(1,1) grey differential equation;

[0092] (4) Finally, the competition coefficients are obtained by estimating the parameters using the least squares method.

[0093] When the parameter inversion method is selected as the Bayesian inference method, it specifically includes:

[0094] (1) First, set a prior probability distribution for the competition coefficients to be solved;

[0095] (2) Establish a likelihood function based on historical water consumption data;

[0096] (3) Using the Markov chain Monte Carlo sampling method, sampling is performed from the posterior probability distribution;

[0097] (4) The posterior mean or the maximum posterior probability estimate calculated based on the sampling results is used as the final estimate.

[0098] When the parameter inversion method is chosen as a genetic algorithm, it specifically includes:

[0099] (1) Define the mean square error between the simulated water consumption and the historical actual water consumption as the fitness function;

[0100] (2) Initialization: Randomly generate a set of candidate solutions, each solution being an encoding of a set of competing coefficients;

[0101] (3) Selection: Based on the fitness function value, select excellent candidate solutions to enter the next generation;

[0102] (4) Crossover and mutation: Perform crossover and mutation operations on the selected candidate solutions to generate new candidate solutions;

[0103] (5) Iterate until the termination condition is met, and take the candidate decoder with the highest fitness in the final generation as the solution of the competition coefficient.

[0104] When the parameter inversion method is selected as a machine learning regression algorithm, it specifically includes:

[0105] (1) Characteristic projects: using historical water consumption, water demand, water supply, and socio-economic indicators of each region as characteristics;

[0106] (2) Label construction: The competition coefficients to be inverted are used as target labels;

[0107] (3) Model training: Use support vector machine, random forest or gradient boosting tree regression algorithm to train the feature to target label mapping model;

[0108] (4) Coefficient estimation: Input new or missing data into the trained model to directly predict the corresponding competition coefficient values.

[0109] S6. Dynamically Quantifying the Competitive Landscape:

[0110] Based on the water resource competition coefficient and the set levels of competition and cooperation intensity, the impact degree, direction and ecological relationship type of water resource competition between regions are dynamically quantified, and the water resource competition pattern between watershed regions is systematically analyzed. The water resource competition coefficient is a metric parameter that quantifies the degree of mutual influence between regions due to competition for limited water resources. Based on the solved water resource competition coefficient, a competition coefficient matrix is ​​constructed.

[0111] ;

[0112] In the formula: The elements in the competition coefficient matrix represent the degree of influence of the j-th region on the i-th region;

[0113] Based on the magnitude and sign of each competition coefficient value in the competition coefficient matrix, and referring to the preset competition and cooperation intensity grading standard table, the water resource relationship between any two regions is divided into specific qualitative levels. At the same time, based on the sign and numerical range of the competition coefficient values, they are mapped to the defined ecological relationship type. By combining the levels and relationship types of all regional pairs, the competitive pattern of water resources in the entire region is systematically analyzed and clarified.

[0114] In the competition and cooperation intensity grading standard table (as shown in Table 1), the relationships are divided into competitive, cooperative, and no relationship based on the different intervals of the competition coefficient. When the competition coefficient is in (-∞, 0), it is a cooperative relationship; when the competition coefficient is in (0, +∞), it is a competitive relationship; when the competition coefficient is equal to 0, it means that there is neither a competitive nor a cooperative relationship between the two regions. Among them, in the cooperative relationship, (-∞, -2] is set as deep cooperation, (-2, -1.5] is set as heavy cooperation, (-1.5, -1] is set as moderate cooperation, and (-1, 0) is set as light cooperation. In the competitive relationship, (0, 1) is set as light competition, [1, 1.5) is set as moderate competition, [1.5, 2) is set as heavy competition, and [2, +∞) is set as fierce competition.

[0115] Table 1. Grading Standards for Competition and Cooperation Intensity Corresponding to Inter-regional Competition Coefficient Values

[0116]

[0117] The ecological relationships corresponding to the defined inter-regional competition coefficient values ​​are shown in Table 2 below:

[0118] Table 2 Ecological Relationships Corresponding to Inter-regional Competition Coefficient Values

[0119]

[0120] The definition rules for ecological relationship types are as follows:

[0121] Using the competition coefficient value β between the i-th region and the j-th region, and the competition coefficient value α between the j-th region and the i-th region, a planar coordinate system is constructed.

[0122] When the planar coordinates are in the first quadrant, that is, both α and β are positive, the ecological relationship between the i-th region and the j-th region is a competitive relationship;

[0123] When the planar coordinates are in the second quadrant, that is, α is negative and β is positive, where β being positive means that the i-th region is in competition with the j-th region, and α being negative means that the j-th region is in cooperation with the i-th region. Therefore, it is inferred that the ecological relationship between the i-th region and the j-th region is a parasitic relationship, where the i-th region is the dominant party and the j-th region is the damaging party.

[0124] When the planar coordinates are in the third quadrant, that is, α is negative and β is negative, where β being negative means that the i-th region has a cooperative relationship with the j-th region, and α being negative means that the j-th region has a cooperative relationship with the i-th region. Therefore, it is inferred that the ecological relationship between the i-th region and the j-th region is a symbiotic relationship.

[0125] When the planar coordinates are in the fourth quadrant, i.e., α is positive and β is negative, where negative β indicates a cooperative relationship between region i and region j, and positive α indicates a competitive relationship between region j and region i. Therefore, it can be inferred that the ecological relationship between region i and region j is a parasitic relationship; where region i is the damaging party and region j is the dominant party.

[0126] When the planar coordinates are on the ordinate axis between the first and second quadrants, i.e. β is a positive number, the ecological relationship between the i-th region and the j-th region is a partial harm-symbiotic relationship.

[0127] When the planar coordinates lie on the ordinate axis between the third and fourth quadrants, β is negative. The ecological relationship between the i-th and j-th regions is a symbiotic relationship.

[0128] When the planar coordinates lie on the horizontal axis between the first and fourth quadrants, α is a positive number. The ecological relationship between the i-th and j-th regions is a symbiotic relationship of partial harm.

[0129] When the planar coordinates are on the horizontal axis between the second and third quadrants, i.e. α is negative, the ecological relationship between the i-th region and the j-th region is a symbiotic relationship.

[0130] When the planar coordinates are at the origin, that is, when both α and β are 0, the ecological relationship between the i-th region and the j-th region is a neutral relationship.

[0131] S7. Scenario Prediction and Risk Warning:

[0132] Different future scenarios are set, and the driving parameters of each scenario are input into a calibrated multi-regional water resource competition differential dynamics model. The model simulates and predicts the development trend of water resource competition under different future scenarios. When the predicted value of the competition coefficient reaches or exceeds a preset threshold, a corresponding level of risk warning is triggered.

[0133] The designed future scenarios include: a baseline scenario, a scenario of rapid economic growth, a scenario of climate change and drought, and a scenario of strict water conservation policies;

[0134] By conducting multi-scenario simulations and threshold judgments, we can achieve a leap from precise quantification to risk level early warning.

[0135] The following are specific implementation examples:

[0136] S1. Define the areas of competition for water use:

[0137] Based on administrative divisions, the five major cities of Heyuan, Huizhou, Dongguan, Guangzhou, and Shenzhen within the Dongjiang River Basin are identified as the regions participating in water resource competition; based on departmental divisions, the four major water-using sectors of each city—agriculture, industry, domestic use, and ecology—are identified as the main competitors.

[0138] S2. Data Acquisition and Preprocessing from Multiple Sources: This process includes two stages: data acquisition and data preprocessing.

[0139] Data acquisition refers to obtaining multi-source data from various sources over the years required for the research through multiple channels;

[0140] In this embodiment, the main data sources include, but are not limited to: obtaining authoritative data on water supply, water consumption, population, and economic development from publicly released "Water Resources Bulletins" and "Statistical Yearbooks" by national and local government departments at all levels; consulting national, provincial, and municipal statistical yearbooks to obtain coherent and standardized socio-economic historical data; and obtaining data on geographical location, land use, digital elevation models, and environmental impact factors such as precipitation obtained through remote sensing inversion from geospatial data cloud platforms and similar geographic information databases for various regions within the basin.

[0141] Data preprocessing involves cleaning and organizing the collected raw data to ensure its integrity and accuracy. The specific processing steps are as follows:

[0142] Data cleaning: Using program scripts or data processing software to identify and delete obviously invalid or erroneous data records, such as abnormal records with negative water supply or water consumption far exceeding the reasonable range;

[0143] Handling missing values: For cases with a small number of missing data, appropriate filling methods such as linear interpolation, mean filling, or spatial interpolation based on neighboring data are used to fill the missing values ​​according to the data characteristics.

[0144] Handling outliers: Employ statistical methods (such as the 3σ principle) or threshold judgment methods based on domain knowledge to identify potential outliers, and review, correct, or remove them by comparing them with historical data or verifying the original data;

[0145] After the above preprocessing steps, a complete, consistent, and accurate normalized time series dataset is finally formed, which can be used for all subsequent calculations and modeling analyses.

[0146] S3. Identify key competitive regions:

[0147] Using preprocessed historical water supply and demand data for each water-using region, the annual water shortage rate of each city is calculated and analyzed. Based on the annual water shortage rate of each city, the water competition intensity between any two cities is calculated. Based on the calculation results of the water competition intensity between any two cities, a historical 5×5 competition intensity matrix is ​​constructed. The competition intensity matrix is ​​analyzed to identify key competitive region pairs.

[0148] S4. Construct a multi-regional water resource competition model:

[0149] Based on the theory of ecological population dynamics, a differential dynamics model of water resource competition among five cities is constructed. Considering the finiteness of water resources, the growth of water use by regional stakeholders, and factors influencing mutual competition, the dynamic changes in the water resource competition relationship among the five cities are analyzed. Specifically:

[0150] Based on the theory of ecological population competition, the variables and parameters in the model are determined, including the annual water consumption, annual water consumption growth rate, and environmental carrying capacity of each city.

[0151] Construct a differential dynamic model of the water resource competition relationship among five cities: its system of differential equations is as follows:

[0152] ;

[0153] In the formula: , representing five cities respectively, and satisfying the following constraints:

[0154] ;

[0155] S5. Inversion solution for competition coefficients:

[0156] The parameters of a water resource competition model for five cities are solved using a parameter inversion method. This parameter inversion method can be selected from multiple grey system theory methods, Bayesian inference methods, genetic algorithms, or machine learning regression algorithms. Those skilled in the art can choose an appropriate method based on data conditions and accuracy requirements. This invention uses grey system theory as an example for the solution.

[0157] Grey system theory is mainly used to describe data systems with incomplete information or randomness. If a system has fuzziness in hierarchical and structural relationships, randomness in dynamic changes, and incompleteness in indicator data, it is called a grey system. These characteristics are called greyness. The core idea of ​​grey system theory is to establish a dynamic model describing the behavior of a grey system based on a small amount of incomplete data, namely grey models (GM models). In a grey system, the original data (0) is transformed into a new data sequence (1) through an accumulation generation method to weaken the randomness of the original data, thereby revealing the law of the system more clearly. Based on the generated and transformed sequence, a differential equation model is established for prediction and description. The specific solution steps are as follows:

[0158] Step 1: Simplify the complex water resource competition model among five cities into a multi-city water resource competition model for quantification. The following is the solution process for one of the simplified inter-city water resource competition models:

[0159] ;

[0160] in,

[0161] ;

[0162] Step 2: Estimate model parameters using the direct modeling method based on grey system theory. Let the original sequence be:

[0163] ;

[0164] Step 3: Generate the original sequence and The first-order cumulative sequence is:

[0165] ;

[0166] Step 4: At time k, define and The gray derivative is:

[0167] ;

[0168] Step 5: Let and For sequence and The continuous average generating sequence, i.e.:

[0169] ;

[0170] Generally taken Substituting into the equation, we get:

[0171] ;

[0172] Step 6: Convert the continuous-time competition model between regions into a discrete-time competition model between regions:

[0173] ;

[0174] Step 7: Represent the discrete competition equations in matrix-vector form as follows:

[0175] ;

[0176] in,

[0177] ;

[0178] ;

[0179] ;

[0180] ;

[0181] Step 8: Solve using linear regression to obtain... The estimated value is:

[0182] ;

[0183] Parameter calibration and model validation: This is used to calibrate and validate the parameters of the solved model to ensure its reliability and robustness. Specifically, it includes:

[0184] (1) Parameter calibration: The model parameters are adjusted using the least squares method, and the mean square error index is used to evaluate the difference between the predicted value and the actual observed data.

[0185] (2) Model Validation: K-fold cross-validation was used for model validation. The historical dataset was randomly divided into K subsets of equal size (e.g., K=5). K-1 subsets were used as the training set for parameter inversion, and the remaining subset was used as the test set for validation. This process was repeated K times. Finally, the predictive performance of the model on different test sets was comprehensively evaluated to measure its generalization ability.

[0186] Sensitivity analysis: used to assess the impact of fluctuations in key parameters (such as growth rate and environmental carrying capacity) on model outputs (such as competition coefficients), specifically including:

[0187] (1) Select analysis parameters: Determine the key parameters that have a significant impact on the model output.

[0188] (2) Set the range of parameter variation: Based on the statistical distribution of the parameters, set a reasonable range of variation for each parameter (such as ±10% or ±20% of the baseline value).

[0189] (3) Perform sensitivity analysis: Use univariate sensitivity analysis, fix other parameters and change only one parameter to observe its impact on the model output; use multivariate sensitivity analysis, change multiple parameters at the same time and observe their combined impact on the model output.

[0190] Single-factor analysis (changing only one parameter at a time) or global sensitivity analysis (changing multiple parameters simultaneously) can be used to observe and record the changes in the model output.

[0191] (4) Record and analyze the results: Record the changes in the model output after each parameter change, and use the sensitivity index to quantify the impact of parameter changes on the model output.

[0192] (5) Conclusion: Based on the results of the sensitivity analysis, the specific parameters are most sensitive to the model output.

[0193] S6. Dynamically Quantifying the Competitive Landscape:

[0194] Based on all the solved competition coefficients, a competition coefficient matrix is ​​constructed. According to the competition and cooperation intensity level classification standard set by this invention, the influence degree of each pair of regional competition relationships is dynamically quantified, the direction of competition influence and its corresponding ecological relationship type are determined, and the water resource competition pattern between watershed regions is systematically analyzed.

[0195] S7. Scenario Prediction and Risk Warning:

[0196] Four future development scenarios were set up: the baseline scenario (developing according to the current trend), the high-speed economic growth scenario (GDP annual growth rate increases to 8%), the climate change drought scenario (average annual rainfall decreases by 20%), and the strict water conservation policy scenario (water-saving technologies improve water use efficiency by 10%). The key driving parameters of these scenarios were input into the calibrated model and the simulation was run until 2035.

[0197] For example, simulation results show that under a climate change drought scenario, the level of competition between regions is expected to exceed the preset "high-risk" threshold around 2030, with a simulated competition coefficient reaching 2.2. The model will automatically trigger a red risk warning, prompting water resource managers to plan ahead and activate inter-regional water resource emergency allocation plans to prevent potential water use conflicts. Under the baseline and economic development scenarios, although the level of competition increases, it does not exceed the high-risk threshold. Based on this, the system issues lower-level blue or yellow warnings to guide routine resource allocation management. Under a strict water conservation policy scenario, the growth trend of the competition coefficient between regions is significantly curbed, with predicted values ​​generally remaining below 1.0 within the range of mild competition or cooperation, and no cases exceeding the set threshold have occurred. Therefore, the system continuously issues green (safe) warnings, indicating that the implemented water conservation policies have effectively alleviated inter-regional competitive pressure, providing managers with positive quantitative evidence for evaluating policy effectiveness and formulating long-term water conservation strategies.

[0198] Example 2

[0199] The present invention provides a dynamic quantification and risk early warning system for multi-regional water resource competition in a river basin, comprising:

[0200] Region and sector identification module: Based on administrative divisions, identify the regions within the basin that participate in water resource competition, and based on sectoral divisions, identify the agricultural, industrial, domestic, and ecological water use sectors that participate in water resource competition within the region;

[0201] Data Acquisition and Preprocessing Module: Collects historical data on water supply, water consumption, population, and economic development level for each region, as well as data on geographical location, climate change, and environmental impact of renewable water resources, and preprocesses the data.

[0202] Competition Intensity Matrix Construction and Key Region Pair Identification Module: Calculates the water shortage rate of each region, and calculates the water resource competition intensity between any two regions based on the calculated water shortage rate of each region, thereby constructing a competition intensity matrix. The constructed competition intensity matrix is ​​then analyzed to identify key competitive region pairs.

[0203] Differential dynamics model construction module: Based on the theory of ecological population competition dynamics, construct a multi-regional differential dynamics model of water resource competition;

[0204] Competition coefficient solution module: The competition coefficient in the multi-region water resource competition differential dynamics model is solved using the parameter inversion method;

[0205] Competition Pattern Analysis Module: Based on the competition coefficient of water resources and the set competition and cooperation intensity levels, the module dynamically quantifies the degree of impact, direction and ecological relationship type of competition between regions, and systematically analyzes the water resource competition pattern between watershed regions.

[0206] Scenario Simulation and Risk Warning Module: Set different future scenarios, input the driving parameters of different scenarios into the calibrated model, simulate and predict the development trend of water resource competition under different future scenarios, and trigger the corresponding level of risk warning when the predicted value of the competition coefficient reaches or exceeds the preset threshold.

[0207] The above are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above content. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for dynamic quantification and risk warning of basin multi-region water resource competition relationship, characterized in that, The method comprises the following steps: S1: According to the administrative division, the regions participating in the water resource competition in the basin are determined, and according to the department division, the agricultural, industrial, domestic and ecological water consumption departments participating in the water resource competition in the region are determined; S2: Collecting the historical data of water supply, water consumption, population, economic development level, geographical location, climate change, renewable water resources and environmental impact data of each region, and preprocessing the data; S3: Calculate the water shortage rate of each region, and calculate the water resource competition intensity between any two regions based on the calculated water shortage rate of each region to construct a competition intensity matrix, analyze the constructed competition intensity matrix, and identify the key competition region pair; S4: Based on the ecological population competition dynamics theory, a multi-region water resource competition differential dynamics model is constructed; S5: The competition coefficient in the multi-region water resource competition differential dynamics model is solved by using a parameter inversion method; S6: According to the competition coefficient of water resources, and based on the set competition and cooperation intensity level, the influence degree, direction and ecological relationship type of the water resource competition between regions are dynamically quantified, and the water resource competition pattern between regions in the basin is systematically analyzed; S7, set different future scenarios, input the driving parameters of different scenarios into the calibrated multi-region water resource competition differential dynamics model, simulate and predict the development trend of water resource competition relationship under different future scenarios, and when the predicted value of the competition coefficient is greater than or equal to the preset threshold, the risk warning of the corresponding level is triggered.

2. The method according to claim 1, wherein, In step S2, the preprocessing includes data cleaning and processing of missing values and outliers.

3. The method according to claim 2, wherein, In step S3, the competition intensity matrix is calculated as follows: The water shortage rate of the region is obtained by calculating the ratio of the difference between the annual water demand and the annual water supply of the region to the annual water supply of the region; The water consumption competition intensity between any two regions is obtained by calculating the first absolute value of the difference between the water shortage rates of the two regions and the second absolute value of the sum of the water shortage rates of the two regions, and taking the ratio of the first absolute value and the second absolute value as the water consumption competition intensity between the two regions; Based on the water consumption competition intensity between any two regions, a competition intensity matrix is constructed: ； In the formula: is an element in the competition intensity matrix, indicating the intensity of mutual influence between the ith region and the jth region, is the competition intensity matrix.

4. The method according to claim 3, characterized in that, In step S4, the constructed multi-region water resource competition differential dynamics model is: ； wherein: is the water use of the i-th region at time t, is the annual growth rate of the i-th region at time t, is the competition coefficient of the j-th region to the i-th region at time t, is the maximum allowed water use of the i-th region at time t, i.e. the environmental carrying capacity, n is the total number of regions, and t is the time in years.

5. The method according to claim 4, wherein, In step S5, the parameter inversion method is one or more of the gray system theory method, the Bayesian inference method, the genetic algorithm or the machine learning regression algorithm.

6. The method according to claim 5, characterized in that, In step S6, based on the solved competition coefficient, a competition coefficient matrix is constructed; According to the size and sign of each competition coefficient in the competition coefficient matrix, the water resource relationship between any two regions is divided into specific qualitative grades by comparing with the preset competition and synergy strength grading standard table, and the sign and value range of the competition coefficient is corresponded to the defined ecological relationship type; the competition pattern of the global water resources is systematically analyzed and clarified by comprehensively considering the grades and relationship types of all region pairs; wherein, the competition coefficient matrix is constructed ; ； In the formula: is a competition coefficient value in the competition coefficient matrix, indicating the influence degree of the jth region on the ith region.

7. The method according to claim 6, wherein, In the competition and cooperation intensity classification standard table, the competition coefficient is divided into competition relationship, cooperation relationship and no relationship according to different intervals; when the competition coefficient is (-∞, 0), it is a cooperative relationship; when the competition coefficient is (0, +∞), it is a competitive relationship; when the competition coefficient is equal to 0, it means that there is no competition relationship and no cooperation relationship between the two regions; wherein, In the synergistic relationship, set (-∞, -2] as deep synergy, set (-2, -1.5] as heavy synergy, set (-1.5, -1] as moderate synergy, and set (-1, 0] as light synergy; in the competitive relationship, set (0, 1] as light competition, set [1, 1.5) as moderate competition, set [1.5, 2) as heavy competition, and set [2, +∞) as intense competition.

8. The method according to claim 7, wherein, The definition rule of the ecological relationship type is: Take the competition coefficient value of the ith region to the jth region as β, and the competition coefficient value of the jth region to the ith region as α, and construct a plane coordinate in this way; When the plane coordinate is in the first quadrant, that is, both α and β are positive numbers, the ecological relationship type between the ith region and the jth region is competitive relationship; When the plane coordinate is in the second quadrant, that is, α is negative and β is positive, where β is positive, representing that the ith region is in competition with the jth region, and α is negative, representing that the jth region is in synergy with the ith region, so it is inferred that the ecological relationship type between the ith region and the jth region is parasitic relationship, where the ith region is the dominant side and the jth region is the damaged side; When the plane coordinate is in the third quadrant, that is, α is negative and β is negative, where β is negative, representing that the ith region is in synergy with the jth region, and α is negative, representing that the jth region is in synergy with the ith region, so it is inferred that the ecological relationship type between the ith region and the jth region is symbiotic relationship; When the plane coordinate is in the fourth quadrant, that is, α is positive and β is negative, where β is negative, representing that the ith region is in synergy with the jth region, and α is positive, representing that the jth region is in competition with the ith region, so it is inferred that the ecological relationship type between the ith region and the jth region is parasitic relationship; where the ith region is the damaged side and the jth region is the dominant side; When the plane coordinate is on the vertical coordinate axis between the first quadrant and the second quadrant, that is, β is positive, the ecological relationship type between the ith region and the jth region is parasitic symbiotic relationship; When the plane coordinate is on the vertical coordinate axis between the third quadrant and the fourth quadrant, that is, β is negative. The ecological relationship type between the ith region and the jth region is partial symbiotic relationship; When the plane coordinate is on the horizontal coordinate axis between the first quadrant and the fourth quadrant, that is, α is positive. The ecological relationship type between the ith region and the jth region is parasitic symbiotic relationship; When the plane coordinate is on the horizontal coordinate axis between the second quadrant and the third quadrant, that is, α is negative, the ecological relationship type between the ith region and the jth region is partial symbiotic relationship; When the plane coordinate is at the coordinate origin, that is, both α and β are 0, the ecological relationship type between the ith region and the jth region is neutral relationship.

9. The method of claim 1, wherein, In step S7, the designed future scenarios include the baseline scenario, the economic high-speed development scenario, the climate change drought scenario, and the strict water saving policy scenario.

10. A system for dynamic quantification and risk early warning of basin multi-region water resource competition relationship, characterized in that, It includes: Region and department identification module: According to the administrative division, the regions participating in the competition for water resources in the basin are determined, and according to the department division, the agricultural, industrial, domestic and ecological water departments participating in the competition for water resources in the region are determined; Data collection and preprocessing module: Collect the historical data of water supply, water consumption, population, economic development level, geographical location, climate change, renewable water resources and environmental impact data of each region, and preprocess the data; Competition intensity matrix construction and key region pair identification module: Calculate the water shortage rate of each region, and calculate the water resource competition intensity between any two regions based on the calculated water shortage rate of each region to construct the competition intensity matrix, analyze the constructed competition intensity matrix, and identify the key competition region pair; Differential dynamics model construction module: Based on the ecological population competition dynamics theory, a multi-region water resource competition differential dynamics model is constructed; Competition coefficient solving module: The competition coefficient in the multi-region water resource competition differential dynamics model is solved by using the parameter inversion method; Competition pattern analysis module: According to the competition coefficient of water resources and based on the set competition and cooperation intensity level, the influence degree, direction and ecological relationship type of the competition between regions are dynamically quantified, and the water resource competition pattern between the regions in the basin is systematically analyzed; Scenario simulation and risk early warning module: Different scenarios are set, the driving parameters of different scenarios are input into the calibrated model, the development trend of water resource competition relationship under different scenarios in the future is simulated and predicted, and when the predicted value of the competition coefficient reaches or exceeds the preset threshold, the risk warning of the corresponding level is triggered.

Images