A method and system for evaluating water-saving effect of a water diversion project receiving area

Abstract

The application relates to the technical field of water resource management, and discloses a method and system for evaluating water-saving effects of a water receiving area of a water diversion project. The method comprises the following steps: collecting historical water consumption and water diversion data of the water receiving area, and constructing water consumption structure distribution information; establishing a water diversion quantity and local water resource coupling model based on the information, and calculating a substitution effect coefficient; based on the coefficient, combining water consumption efficiency benchmarks and water-saving potential, constructing a multi-industry water-saving effect hierarchical evaluation system; identifying industry water-saving abnormalities through the system, attributing to water right transaction data to obtain a contribution degree ranking; based on the abnormal labels and the ranking, constructing a water-saving effect time series model, and extracting time evolution characteristic parameters; according to the parameters and spatial distribution data, constructing a time-space evolution analysis model, and generating a time-space distribution graph of water-saving influence contribution values; based on the graph, formulating a water resource allocation scheme, and generating an evaluation report. The application effectively improves the fine level of water resource allocation.

Description

Technical Field

[0001] This application relates to the field of water resources management technology, and in particular to a method and system for evaluating the water-saving effectiveness of water diversion projects in water-receiving areas. Background Technology

[0002] In water resource management within the water-receiving areas of water diversion projects, scientifically evaluating water-saving effectiveness is fundamental to optimizing inter-basin water allocation. Currently, water-receiving areas generally experience a mixed supply of local water sources and diverted water, while water demand, water intake structures, and water-saving potential vary significantly across industries. Existing evaluation methods are mostly based on static changes in total water consumption or single water-saving indicators, lacking dynamic quantification of the local water source response process after diverted water intervention. This makes it difficult to accurately identify the actual substitution effect of diverted water on local water resources. This deficiency results in evaluation results that fail to truly reflect the real water-saving contributions of water diversion projects across industries and time periods, hindering the development of differentiated water resource allocation strategies.

[0003] Furthermore, the water users across various industries in the water-receiving area vary significantly in their water-saving technology levels and water use efficiency, and water rights are often redistributed through water rights trading. This makes the attribution of water-saving achievements increasingly complex. Traditional methods often attribute the overall regional water-saving effect broadly to water diversion projects or distribute it equally among various water-using industries, failing to establish a mapping relationship between substitution effects, water use efficiency benchmarks, and water rights transfer. When some industries' insufficient water-saving efforts are masked by the transfer of water rights, or when the contributions of highly efficient water-saving industries to the transfer of water rights are not fully recognized, the evaluation system becomes ineffective. It cannot accurately pinpoint the source and contributor of abnormal water-saving practices, thus failing to provide an effective basis for the refined regulation of water resources within the region.

[0004] Furthermore, the aforementioned substitution effects and water-saving contributions exhibit temporal cumulativeity and spatial heterogeneity. The effect of water diversion on reducing local water consumption varies with time lags and fluctuations across different time windows. The geographically dispersed distribution of various water users further contributes to uneven spatial distribution of water-saving effectiveness. Existing technological solutions largely rely on macro-level statistics of time series or static benchmarking of single points, lacking the ability to integrate and analyze industry-scale temporal evolution parameters with spatial distribution data. This fails to generate visual evidence reflecting the spatiotemporal dynamic distribution of water-saving impacts. Consequently, configuration schemes targeting specific water-scarce regions lack scenario simulation and predictive validation, making it difficult to support the spatiotemporal balanced allocation of water resources and long-term water-saving management in water-receiving areas. Summary of the Invention

[0005] This application provides a method and system for evaluating the water-saving effectiveness of water diversion projects in the water-receiving area, which can be used to improve the precision of water resource allocation.

[0006] Firstly, this application provides a method for evaluating the water-saving effectiveness of a water diversion project in the receiving area, including: S1. Collect historical water usage data and water diversion and supply records of various industries in the water-receiving area to construct information on the distribution of water usage structure in the water-receiving area; S2. Based on the water use structure distribution information of the water receiving area, establish a dynamic coupling model between the diverted water volume and local water resources, and calculate the substitution effect coefficient of the diverted water on local water resources. S3. Based on the substitution effect coefficient, and combined with the water use efficiency benchmark and water-saving potential coefficient of various industries in the water-receiving area, construct a hierarchical evaluation index system for water-saving performance of multiple industries in the water-receiving area. S4. Through the water-saving effectiveness hierarchical evaluation index system, identify water-saving anomalies in various industries in the water-receiving area, obtain water-saving anomaly labels, and combine water rights trading data to attribute water-saving contributions and obtain contribution ranking. S5. Based on the water-saving anomaly labels and contribution ranking, construct a time series model of water-saving effectiveness in the water-receiving area, analyze the time evolution law of water-saving effectiveness in the water-receiving area, and form a set of time evolution characteristic parameters. S6. Based on the set of time evolution characteristic parameters and the spatial distribution data of the water-receiving area, construct a spatiotemporal evolution analysis model of the water-saving effect of the water-receiving area, and determine the spatiotemporal distribution map of the water-saving impact contribution value of the water-receiving area. S7. Based on the spatiotemporal distribution map of the water-saving impact contribution value, formulate a water resource allocation plan for the water-receiving area and generate a water-saving effectiveness evaluation report.

[0007] Secondly, this application provides a water-saving performance evaluation system for the water-receiving area of ​​a water diversion project, comprising: The water use structure analysis module is used to collect historical water use data and water diversion and supply records of various industries in the water-receiving area, and to construct water use structure distribution information of the water-receiving area. The substitution effect calculation module is used to establish a dynamic coupling model between the diverted water volume and local water resources based on the water use structure distribution information of the water receiving area, and to calculate the substitution effect coefficient of the diverted water on local water resources. The evaluation system construction module is used to construct a hierarchical evaluation index system for water-saving performance of multiple industries in the water-receiving area based on the substitution effect coefficient, combined with the water use efficiency benchmark value and water-saving potential coefficient of each industry in the water-receiving area. The anomaly attribution identification module is used to identify water-saving anomalies in various industries in the water-receiving area through the water-saving effectiveness hierarchical evaluation index system, obtain water-saving anomaly labels, and combine water rights trading data to attribute water-saving contributions and obtain contribution ranking. The time-series evolution analysis module is used to construct a time-series model of water-saving effectiveness in the water-receiving area based on the water-saving anomaly labels and contribution ranking, analyze the time evolution law of water-saving effectiveness in the water-receiving area, and form a set of time evolution feature parameters. The spatiotemporal evolution modeling module is used to construct a spatiotemporal evolution analysis model of the water-saving effect of the water-receiving area based on the set of time evolution characteristic parameters and the spatial distribution data of the water-receiving area, and to determine the spatiotemporal distribution map of the water-saving impact contribution value of the water-receiving area. The scheme report generation module is used to formulate a water resource allocation scheme for the water-receiving area based on the spatiotemporal distribution map of the water-saving impact contribution value, and generate a water-saving effectiveness evaluation report.

[0008] Thirdly, this application provides a computer device comprising: a memory and at least one processor, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via a bus, and when the machine-readable instructions are executed by the processor, the steps of the above-described method for evaluating the water-saving effectiveness of a water diversion project in a water-receiving area are performed.

[0009] Compared with the prior art, the beneficial effects of this application are at least as follows: 1. By constructing a dynamic coupling model between the diverted water volume and local water resources, the magnitude and time lag characteristics of the reduction in local water consumption are extracted according to industry and time window, and the parameters are dynamically calibrated. This enables the accurate quantification of the water diversion substitution effect under different industries and time periods, solving the technical problem that existing methods cannot effectively separate the true substitution contribution of diverted water.

[0010] 2. By integrating the substitution effect coefficient with water use efficiency benchmark, water-saving potential coefficient and water rights trading data, a mapping relationship between anomaly labels and contribution is established and contribution ranking is achieved, which solves the shortcomings of traditional methods that cannot accurately attribute the real water-saving contribution of various industries under the premise of water rights transfer.

[0011] 3. By introducing water-saving anomaly labels as external intervention variables into the time series model, and combining the changes in the ranking of contribution and the characteristics of distribution concentration for predictive analysis, dynamic modeling and anomaly early warning of the time-series evolution of water-saving effectiveness in various industries can be achieved, making up for the lack of time-series prediction capabilities in existing technologies.

[0012] 4. By fusing temporal evolution characteristic parameters with the spatial attribution matrix of water users, a spatiotemporal distribution map of the contribution value of water-saving impact is generated. Scenario simulation verification is then conducted in water-scarce areas with different causes, providing a systematic solution for the spatiotemporal balanced allocation of water resources. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 A flowchart of a water-saving performance evaluation method for a water diversion project receiving area according to this application; Figure 2 A comparison chart showing the water-saving effects before and after the implementation of the configuration scheme in this application; Figure 3 This is a schematic diagram of the structure of a water-saving performance evaluation system for a water diversion project in the water-receiving area, as described in this application. Figure 4 This is a schematic block diagram of a water-saving performance evaluation device for a water diversion project in the water-receiving area, as described in this application. Detailed Implementation

[0015] This application provides a method and system for evaluating the water-saving effectiveness of a water diversion project in the receiving area. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0016] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of a method for evaluating the water-saving effectiveness of a water diversion project in the receiving area, as described in this application, includes: Step S1: Collect historical water usage data and water diversion and supply records of various industries in the water-receiving area to construct water usage structure distribution information of the water-receiving area.

[0017] In one specific embodiment, the process of performing step S1 may specifically include the following steps: Collect information on water usage time periods, water consumption, and water sources in industrial, agricultural, and domestic water-receiving areas, and integrate them to generate historical water usage data; Collect water supply records for water diversion projects in the water-receiving area. The water supply records include the amount of water diverted, the time period of water diversion, and the allocation information of the water-receiving area. Historical water use data and water supply records are preprocessed to remove outliers, and then aligned and integrated according to time series to construct information on the distribution of water use structure in the water-receiving area.

[0018] Specifically, water use structure distribution information for water-receiving areas is constructed through multi-source data collection and standardized integration. When collecting information on water use periods, consumption, and sources for industrial, agricultural, and domestic water-receiving areas, industrial water use data is obtained from enterprise metering terminals, identifying water use periods and consumption for each production unit, and classifying water sources into three categories—surface water, groundwater, and diverted water—based on water withdrawal permits. Agricultural water use relies on irrigation district ledgers recording the start and end dates of each irrigation cycle as water use periods, using the net irrigation water volume after deducting water transfer losses from the headworks as water consumption, and recording the proportions of reservoir storage, well groundwater, and diverted water from main canals as water source information. Domestic water use data is obtained from the water supply company's billing system, with monthly water consumption data obtained, water use periods for each area determined based on pump station operation logs, and diverted water access status marked in pipeline valve scheduling records as water source information. The data from these three areas are integrated into a unified structure to generate historical water use data, using industry category, timestamp, and source type as a joint primary key to ensure that each record is traceable to a specific industry, time period, and water source.

[0019] When collecting water supply records of water diversion projects in the water receiving area, the water diversion volume for each diversion is obtained from the water diversion project dispatch center, including the total water diversion volume at the head of the canal and the measured water diversion volume at each branch gate. The difference between the two is used to calibrate the water transmission loss rate, providing parameter basis for accurately calculating the actual available net water supply of each industry when constructing the dynamic coupling model in step S2. For example, if the total water diversion volume at the headworks of a certain water diversion canal is 10 million cubic meters, and the sum of the measured water diversion volumes at each branch gate along the canal is 8.5 million cubic meters, the difference of 1.5 million cubic meters is the water transmission loss. The calibrated water transmission loss rate is 15%. When calculating the actual water diversion volume obtained by agriculture, industry, and other sectors, this ratio must be deducted to obtain the effective supply. The opening and closing times of the gates are recorded as the water diversion period, accurate to the hour. Based on the dispatch instructions and water allocation agreements, the proportion or volume of each batch of diverted water allocated to each administrative region and the three sectors of industry, agriculture, and domestic life within that region is recorded as the allocation information for the water-receiving areas.

[0020] When preprocessing historical water use data and water supply records to remove outliers, a sliding window statistical method is used. The window width is 12 months, and the step size is 1 month. The median and absolute median difference of the water consumption series within the window are calculated. Observations that deviate from the median by more than 3 times the absolute median difference are marked as candidate outliers. The candidate outliers are then verified twice using an industry knowledge base. Only records distorted by metering failures, data entry errors, or transmission packet loss are removed, while fluctuations caused by reasonable events such as industrial production restrictions or drought emergency dispatch are retained. For missing segments after outlier removal, gaps of no more than two time windows are filled by multiplying the average water consumption of the same period in the three nearest years by the industry's water use trend factor over the past three years. Gaps exceeding two time windows are filled by regression estimation using external correlation variables such as the industrial output index or the percentage of precipitation anomaly.

[0021] After data cleaning, using months as the unified time granularity, water usage data from various industries were registered and aligned with water diversion and supply records along the time dimension. This generated a two-dimensional data structure with months as row indexes and industry category and source type as column indexes. For each industry, characteristic indicators such as the proportion of diverted water usage to total water usage, the proportion of local water source usage, the range of fluctuations in proportions between months, and the load sharing rate of diverted water during peak water usage months were calculated. These factors were then used to comprehensively construct water usage structure distribution information for the water-receiving area. For example, the water usage structure distribution information constructed for a certain water-receiving area shows that the proportion of diverted water usage in the industrial sector fluctuates between 15% and 25% throughout the year, reaching a peak during the hot summer months due to increased cooling water consumption; the proportion of diverted water usage in the agricultural sector can reach over 50% during peak spring and summer irrigation periods, but is close to 0 during the winter fallow period; the proportion of diverted water usage in the residential sector is relatively stable throughout the year, only slightly increasing during the summer peak water usage period due to insufficient local reservoir storage.

[0022] This water use structure distribution information records the differentiated dependence patterns of different industries on diverted water and local water sources in different seasons, providing standardized data for calculating substitution effect coefficients by industry and time window.

[0023] Step S2: Based on the water use structure distribution information of the water receiving area, establish a dynamic coupling model between the diverted water volume and local water resources, and calculate the substitution effect coefficient of the diverted water on local water resources.

[0024] In one specific embodiment, the process of performing step S2 may specifically include the following steps: Based on the water use structure distribution information of the water receiving area, the local water resource usage before and after the water diversion is statistically analyzed by industry and time window, the local water source reduction of each industry in each time window is calculated, the magnitude characteristics and time lag characteristics of the reduction are extracted, and a set of local water source response change characteristics by industry is formed. Using the amount of water diverted as the independent variable and the amount of local water source reduction as the dependent variable, and utilizing the local water source response change feature set, a substitution relationship curve is fitted for each industry. The substitution relationship curve is dynamically calibrated according to time windows, and a dynamic coupling model of water diversion and local water resources is established. The substitution effect coefficient of each industry under each time window is output through the dynamic coupling model.

[0025] Specifically, a dynamic coupling model of water diversion volume and local water resources is established by combining industry-specific and time-period-specific statistics with distributed lag regression modeling, and the substitution effect coefficient is calculated. Based on the water use structure distribution information of the water-receiving area, the local water resource usage before and after the intervention of water diversion is statistically analyzed by industry and time window. The time window refers to a discrete analysis unit that divides the continuous time series according to water use cycle characteristics such as monthly or crop growing season. The local water source usage is independently statistically analyzed within each time window to capture the response differences under different seasons and water use stages. For each industry in each time window, the time when the diverted water first arrives at the water metering node of the industry is used as the intervention boundary point. The local water source usage is statistically analyzed backward for the same duration as the baseline value, and the actual local water source consumption in the period after intervention is statistically analyzed backward. The reduction of local water source consumption in the industry within the time window is obtained by subtracting the actual consumption from the baseline value. The distributed lag regression model uses the current period and several lag periods of water diversion as independent variables, and the local water source reduction as the dependent variable. The model structure is as follows: the local water source reduction rate equals a constant term, plus the regression coefficient of the current immediate effect multiplied by the current water diversion input ratio, plus the regression coefficient of each lag effect multiplied by the corresponding lag period water diversion input ratio, and finally the random error term is added. The lag order is determined by the response lag determined in the time lag feature extraction, and the regression coefficients of the model are estimated by ordinary least squares method.

[0026] Based on the calculation of water consumption reduction, the magnitude and time lag characteristics of the reduction are extracted to form a set of local water source response change characteristics for each industry. The magnitude characteristic refers to the strength of the change in local water source consumption reduction before and after the intervention of diverted water, including two dimensions: absolute magnitude and relative magnitude. The absolute magnitude is the absolute value of the reduction within the time window, and the relative magnitude is the percentage of the reduction relative to the baseline local water source usage before the intervention of diverted water. Both together characterize the strength of the substitution effect. The time lag characteristic refers to the response delay between the arrival of diverted water at the end of the water user's journey and the actual reduction of local water source usage by the water user. Its extraction is achieved by constructing a cross-correlation function between the diverted water input sequence and the local water source consumption reduction sequence: using the diverted water input sequence as the reference sequence and the local water source consumption reduction sequence as the response sequence, the cross-correlation coefficients under different lag orders are calculated. The lag order corresponding to the maximum value of the cross-correlation coefficient is selected as the response time lag for the industry within that time window. This time lag characterizes the operational cycle required for the water user to complete the switching of water intake facilities or the adjustment of water use plans from receiving diverted water. The amplitude and time delay characteristics mentioned above are organized in a structured manner according to industry and corresponding time window. Each industry corresponds to a time series containing amplitude and time delay values ​​within multiple time windows. The set of series for all industries is the local water source response change characteristic set.

[0027] Using the amount of water diverted as the independent variable and the reduction in local water consumption as the dependent variable, a substitution relationship curve was fitted for each industry using a set of local water consumption response change characteristics. The substitution relationship curve is a regression function describing the quantitative response relationship between the amount of water diverted and the reduction in local water consumption. Before fitting, the variables in the feature set were normalized. The normalization benchmark was the multi-year average of local water consumption before the intervention of water diversion in each industry. By converting the absolute magnitude to a relative reduction rate and the amount of water diverted to a proportion relative to the industry's average annual water consumption, the interference of differences in water consumption among different industries on the estimation of model parameters was eliminated. After normalization, a distributed lag model is independently constructed for each industry for regression fitting. This distributed lag model is a time series regression model that incorporates the current period and several lag periods of the independent variable into the regression equation. The model structure includes the regression coefficients of the immediate effect of water diversion input in the current period and the regression coefficients of the lag effects of water diversion input in each lag period. Its general form is that the local water source reduction rate equals the regression coefficient of the immediate effect multiplied by the proportion of water diversion input in the current period, plus the regression coefficients of each lag effect multiplied by the proportion of water diversion input in the corresponding lag period, plus a random error term. The lag order is determined by the response lag determined in the aforementioned time lag feature extraction process. The model parameters are estimated using ordinary least squares. The normalized sample data of all time windows for each industry are used to construct the design matrix and response vector according to the structure of this distributed lag model. The optimal estimates of each regression coefficient are obtained by minimizing the sum of squared residuals, thereby obtaining the substitution relationship curve for that industry.

[0028] Based on this, the substitution relationship curve is dynamically calibrated according to time windows, establishing a dynamic coupling model between the diverted water volume and local water resources. Dynamic parameter calibration refers to calibrating the parameters of the substitution relationship curve window by window based on actual observation data within different time windows, ensuring that the curve parameters for the same industry in different seasons or water use stages reflect the unique water use behavior characteristics and water supply conditions of that period. The calibration process uses rolling time window regression: a fixed-length rolling window is set, the length of which is determined based on the seasonal periodicity of water use data in the water-receiving area, typically four time windows to cover a complete water use fluctuation cycle; starting from the beginning of the time series, it slides forward with a step size of one time window. At each sliding position, normalized sample data falling within that window is used as the training dataset to re-estimate all parameters of the distributed lag model, thereby generating a set of parameter vectors specific to each time window, containing the regression coefficients of the immediate effects and the regression coefficients of each lag effect within that window. During the rolling calibration process, the goodness of fit of the regression for each window is recorded synchronously to evaluate the stability of the response relationship between water diversion input and local water source reduction within that window. Time windows with a goodness of fit below a preset threshold are marked as abnormal response windows, indicating that external factors may interfere with the normal functioning of the substitution effect during that period. After completing the rolling calibration of all time windows, a parameter matrix is ​​obtained, with industry as the row, time window as the column, and parameter vector as the element. This parameter matrix, together with the distributed lag model structure, constitutes a dynamic coupling model.

[0029] The dynamic coupling model works as follows: it receives the time series of diverted water volume as input data, retrieves the corresponding parameter vector from the parameter matrix according to the index of the target time window, substitutes the regression coefficients of the immediate effect and the regression coefficients of each lagged effect in the parameter vector into the distributed lag model structure, and calculates the substitution effect coefficients for each industry under each time window by combining the diverted water input ratio in the current period and historical lag periods. The substitution effect coefficient is defined as the change in the local water source reduction rate caused by the change in the proportion of diverted water input per unit time window and several historical lag windows. This coefficient integrates the regression coefficients of the immediate effect and the regression coefficients of each lagged effect, fully reflecting the time-cumulative effect of diverted water on the substitution of local water sources. The larger the coefficient value, the more obvious the substitution effect of diverted water on local water sources in that industry within that time window.

[0030] For example, the local water source response change characteristic set of the agricultural sector in a certain water-receiving area shows that the amplitude characteristic of the spring irrigation window is that the local groundwater extraction volume decreases significantly after the diversion of water, with a relatively high proportion reaching the baseline level before the intervention. The time lag characteristic extracted by the cross-correlation function shows that the local well extraction volume only begins to decrease significantly about one month after the diverted water arrives in the canal system, with a response lag of about one time window. The amplitude characteristic of the industrial sector during the summer peak window is also obvious, but the time lag characteristic shows that the response lag is close to zero, and the local water source consumption decreases almost synchronously after the diversion of water. The fitting results of the distributed lag model show that the regression coefficient of the lag effect in the agricultural sector during the spring irrigation window is large, while the regression coefficient of the immediate effect is relatively small, reflecting that the main agricultural water users need to go through the operation cycle of irrigation system adjustment and water intake facility switching. In contrast, the regression coefficient of the immediate effect dominates in the industrial sector, while the regression coefficient of the lag effect is not obvious, reflecting the immediate response characteristics of industrial water dispatch to the arrival of diverted water. After being calibrated with a rolling window, the substitution effect coefficients output by the dynamic coupling model indicate that during the spring irrigation period, the agricultural sector can experience a significant reduction in local water resources per unit of diverted water input, and this effect peaks in the second month of continuous diverted water supply; while the substitution effect coefficients for the industrial sector are relatively stable and have almost no time delay.

[0031] Step S3: Based on the substitution effect coefficient, and combined with the water use efficiency benchmark and water-saving potential coefficient of various industries in the water-receiving area, construct a hierarchical evaluation index system for water-saving performance of multiple industries in the water-receiving area.

[0032] In one specific embodiment, the process of performing step S3 may specifically include the following steps: Obtain water use efficiency benchmarks and water-saving potential coefficients for various industries in the water-receiving area, and multiply the substitution effect coefficient as a weighting adjustment factor with the water use efficiency benchmarks to establish water-saving performance evaluation benchmarks for each industry. The water use efficiency deviation, water conservation contribution compliance rate, and water diversion dependence are extracted from the water conservation performance evaluation benchmark and integrated to generate a multi-dimensional water conservation performance evaluation index set. Based on a multi-dimensional water-saving performance evaluation index set, stratified thresholds for each index are set, and water-saving performance is stratified for various industries in the water-receiving area, thus constructing a stratified evaluation index system for water-saving performance across multiple industries in the water-receiving area.

[0033] Specifically, the water use efficiency benchmark value refers to the benchmark value of water consumption per unit product or irrigation water consumption per unit area, determined according to advanced water use quotas within the industry or the strictest regional water resource management system assessment standards. It represents the reasonable water use level that the industry should achieve under current technological conditions. The water use quota standards promulgated by the province where the water-receiving area is located are collected, and the advanced quota value is taken as the initial benchmark value. Simultaneously, the water use efficiency control targets set for each industry in the strictest regional water resource management system assessment documents are obtained. The initial benchmark value is compared with the control targets, and the stricter one is taken as the water use efficiency benchmark value. The water-saving potential coefficient refers to the proportion of exploitable water-saving space derived by comparing the difference between the industry's actual water consumption and the water use efficiency benchmark value, combined with a comprehensive assessment of the feasibility and economic affordability of water-saving technological transformation in the industry. Its value ranges from 0 to 1. Specifically, the process begins by calculating the percentage of the difference between the current actual unit water consumption and the water efficiency benchmark value in the industry, which represents the theoretical water-saving potential. Then, the feasibility of the industry's registered water-saving technology upgrade plans is assessed, with the assessment dimensions including technology maturity level and equipment replacement cycle. Simultaneously, the industry's average profit margin over the past three years is collected as an economic affordability indicator. The technology feasibility score and the economic affordability indicator are weighted together to form a discount coefficient for implementation difficulty, ranging from 0 to 1. Finally, the theoretical water-saving potential is multiplied by the implementation difficulty discount coefficient to obtain the water-saving potential coefficient. The water-saving performance evaluation benchmark is a scale formed by weighting and adjusting the water use efficiency benchmark value using the substitution effect coefficient as a weighting adjustment factor. The construction process is as follows: First, the water use efficiency benchmark value and the water-saving potential coefficient are normalized to eliminate the inconsistency of dimensions caused by differences in products and scales between different industries. The normalization benchmark adopts the multi-year average of the water use efficiency benchmark value of each industry. The normalization process adopts the minimum-maximum normalization method, which linearly maps the original data to the interval of 0 to 1 by subtracting the minimum value and then dividing by the difference between the maximum and minimum values. Then, the normalized water use efficiency benchmark value is multiplied by the substitution effect coefficient, which is used as a weighting adjustment factor. The larger the coefficient value, the more obvious the effect of water diversion to replace local water sources in the industry. Therefore, the expected value of water-saving performance of the industry is correspondingly increased under the same water use efficiency conditions.

[0034] Water use efficiency deviation refers to the degree of deviation between the industry's actual water use efficiency and the water-saving performance evaluation benchmark. It is calculated as the percentage of the difference between the industry's actual unit water intake and the water use efficiency target value adjusted for the substitution effect coefficient in the water-saving performance evaluation benchmark. This indicator reflects the effectiveness of the industry's own water-saving measures. Water-saving contribution compliance rate refers to the proportion of water saved by the industry within the water-receiving area due to the combined effect of improved water use efficiency and water diversion substitution, relative to the industry's expected water-saving target. The water-saving target is determined comprehensively based on the industry's water-saving potential coefficient and water use scale. A higher compliance rate indicates a more substantial contribution by the industry to overall water-saving effectiveness. The dependence on diverted water refers to the comprehensive evaluation value of the proportion of diverted water in the total water consumption of an industry and its substitution effect coefficient. It is obtained by weighting the proportion of diverted water use and the substitution effect coefficient. The weight allocation uses the normalized value of the substitution effect coefficient as the adjustment weight for the proportion of diverted water use to reflect the differentiated impact of different levels of substitution on the degree of dependence. This indicator reflects the degree to which the industry's water-saving achievements depend on the supply of diverted water; excessively high dependence indicates a risk to the sustainability of the industry's water-saving achievements. The above three indicators are organized into an indicator matrix according to industry and time window. Each industry corresponds to a set of indicator vectors for each time window, including water use efficiency deviation, water-saving contribution achievement rate, and dependence on diverted water. This indicator matrix is ​​the multi-dimensional water-saving performance evaluation indicator set.

[0035] The stratification threshold refers to the critical value for classifying the evaluation levels of each indicator. The stratification threshold is set based on the water resource management objectives of the water-receiving area and the phased requirements of the industry's water conservation plan. Each indicator is divided into three level intervals, corresponding to three states: excellent, compliant, and abnormal. For each industry at each time window, the values ​​of the three indicators in its multi-dimensional water conservation performance evaluation set are compared one by one with the corresponding stratification threshold to determine the level interval of each indicator. Based on the combined level determination results of the three indicators, the industry's water conservation performance is divided into three levels according to the preset stratification rules: priority maintenance level, routine management level, and key rectification level. This forms a stratified evaluation indicator system for water conservation performance across multiple industries in the water-receiving area. This system outputs the water conservation performance level label for each industry at each time window, providing a structured evaluation framework for the accurate identification of water conservation anomalies and the reasonable attribution of water conservation contributions from various industries.

[0036] Step S4: Identify water-saving anomalies in various industries within the water-receiving area through a stratified evaluation index system for water-saving effectiveness, obtain water-saving anomaly labels, and combine water rights trading data to attribute water-saving contributions and obtain a ranking of contributions.

[0037] In one specific embodiment, the process of performing step S4 may specifically include the following steps: Through a hierarchical evaluation index system for water-saving effectiveness, the water-saving performance of various industries in the water-receiving area is evaluated in real time. Industries with water-saving performance below the preset performance threshold are marked as water-saving abnormal industries, and a water-saving abnormal label containing an industry identification code and an abnormal deviation level is generated for each water-saving abnormal industry. Obtain transaction data from the water rights trading platform, extract the water rights transfer volume and direction of water users in various industries, count the water rights transferred out as positive water-saving contribution, and the water rights transferred in as negative water-saving contribution, and calculate the water-saving contribution of each water user. The water-saving contribution is aggregated by industry, and a mapping relationship is established between industry water-saving contribution and water-saving anomaly labels. The water-saving contribution attribution results of the water-receiving area are generated, and the contribution ranking of each water user to the overall water-saving effect of the water-receiving area is output.

[0038] Specifically, driven by a rolling evaluation cycle, a water-saving performance scan of the entire industry is triggered at the end of each time window. The scan retrieves three indicators from the current multi-dimensional water-saving performance evaluation index set: water use efficiency deviation, water-saving contribution compliance rate, and dependence on diverted water. These indicators are then compared item by item with preset performance thresholds. The preset performance thresholds are determined by collecting the phased assessment indicator values ​​set for each industry in the water-saving society construction plan of the watershed or provincial administrative region where the water-receiving area is located over the past five years. The median of the indicator values ​​for each industry is then used as the benchmark threshold after grouping them by industry category. If the above planning data is unavailable, the default performance thresholds are set as follows: water use efficiency deviation not exceeding 20%, water-saving contribution compliance rate not less than 80%, and dependence on diverted water not exceeding 70%. The preset performance thresholds correspond to the boundary criteria between the regular management level and the key rectification level in the tiered evaluation index system for water-saving performance. If any indicator of any industry falls within the key rectification level range, that industry is marked as an abnormal water-saving industry. After the marking is completed, the system automatically generates a water-saving anomaly label for each water-saving industry with an anomaly. The data structure of this label includes two fields: industry identification code and anomaly deviation level. The industry identification code adopts the industry classification standard coding of the water-receiving area. The anomaly deviation level is divided into three levels: slight deviation, moderate deviation, and severe deviation, based on the deviation of the indicator value from the threshold. Slight deviation is when the indicator value deviates from the threshold by less than 10%, moderate deviation is when the deviation is between 10% and 30%, and severe deviation is when the deviation exceeds 30%. The larger the deviation, the higher the anomaly deviation level.

[0039] While marking anomalies, transaction data is retrieved from the water rights trading platform's data interface to extract the water rights transfer volume and direction for each industry's water users. Water rights transfer volume refers to the number of water rights quota changes completed by water users on the trading platform through negotiated transfers or auctions. Transfer direction is divided into two types: transfer out and transfer in. Based on the extraction results, water rights transfer out is counted as a positive water-saving contribution, and water rights transfer in is counted as a negative water-saving contribution. The water-saving contribution rate for each water user is calculated using the formula: the water-saving contribution rate equals the percentage of the net value of water rights transfer out minus water rights transfer in relative to the total water withdrawal permit of that water user. For water users who did not participate in water rights trading, their water-saving contribution rate is directly taken as the inverse value of their water use efficiency deviation after normalization; the lower the water use efficiency deviation, the higher the water-saving contribution rate.

[0040] Water-saving contributions are aggregated by industry, using industry identifiers as the aggregation key. The water-saving contributions of each water user within the same industry are weighted and summed, with the weights calculated as the proportion of each water user's water consumption to the industry's total water consumption. The preset contribution threshold is determined as follows: All industries in the water-receiving area are sorted from highest to lowest water-saving contribution. The lowest contribution value corresponding to the top-ranked industries is used as the initial threshold. If no historical ranking data is available, a default preset contribution threshold is set where the industry's water-saving contribution is greater than zero and not less than 60% of the industry average. A mapping relationship is established between industry water-saving contributions and water-saving anomaly labels. The mapping rule cross-correlates industry water-saving contributions with anomaly deviation levels. If an industry's water-saving contribution is below the preset contribution threshold and the industry has been labeled as having water-saving anomalies, the industry is classified as having internal water-saving insufficiency anomalies. If an industry's water-saving contribution is above the preset contribution threshold but still has water-saving anomaly labels, it is classified as having insufficient water supply or water efficiency benchmark deviation anomalies.

[0041] Step S5: Based on the water-saving anomaly labels and contribution ranking, construct a time series model of water-saving effectiveness in the water-receiving area, analyze the time evolution law of water-saving effectiveness in the water-receiving area, and form a set of time evolution characteristic parameters.

[0042] In one specific embodiment, the process of performing step S5 may specifically include the following steps: Extract the ranking and distribution concentration features of the contribution ranking by time window to form a time series feature vector of the water-saving contribution structure of the water-receiving area. The occurrence time and duration of water-saving anomaly labels are used as external intervention variables. Time series feature vectors are introduced, and a time series prediction algorithm with exogenous variables is used to construct a time series model of water-saving effectiveness in the water-receiving area. By using a time series model of water-saving effectiveness, the water-saving effectiveness of the water-receiving area in each future time window is predicted and analyzed, and the predicted trend curves and abnormal early warning signals of water-saving effectiveness in various industries are obtained, which are then compiled into a set of time evolution characteristic parameters.

[0043] Specifically, using the contribution ranking table as input, feature extraction is performed for each time window. During extraction, the ranking changes of all water users between adjacent windows are iterated, and the mean and variance of the ranking differences between adjacent windows for each user are calculated as two components of the ranking change feature. Simultaneously, the sum of squares of the contribution share of each industry is calculated to obtain the Herfindahl index, which serves as the distribution concentration feature. These three statistical components are arranged in window order to form a time-series feature vector of the water-saving contribution structure of the water-receiving area.

[0044] During the intervention variable construction phase, the system scans each time window for water-saving anomaly labels. If a label is found, the intervention variable for that window is assigned the product of the numerical code for the anomaly deviation level and the duration. The anomaly deviation level is coded as 1, 2, and 3 for mild, moderate, and severe, respectively. The duration is measured by the number of consecutive windows in which the anomaly label appears. If no anomaly label is found, the intervention variable is set to zero. The intervention variable sequence is then merged with the time series feature vector, aligned by window.

[0045] In the modeling phase, a vector autoregressive model with exogenous variables was selected. This model treats the three components of the time series feature vector as endogenous variables and the intervention variable as exogenous variables, describing the dynamic dependencies between the endogenous variables and between them and the exogenous intervention through a set of simultaneous equations. The model parameters to be estimated include the regression coefficient matrix of each endogenous variable with respect to all lagged terms of endogenous variables, the influence coefficient vector of the exogenous variable on the current and lagged terms of each endogenous variable, and the covariance matrix of the random disturbance terms of the three endogenous variables. The lag order is automatically selected from a candidate range of 1 to 8 using the Bayesian information criterion. The model parameters are jointly estimated using the generalized least squares method through a seemingly uncorrelated regression approach, iteratively solved until the covariance matrix converges.

[0046] The training data is divided using a time-series cross-validation strategy, with approximately 70% of the total time window used as the initial training set and the remaining 30% as the test set. During training, the design matrix is ​​constructed and all parameters are estimated using the training set. During prediction, a single-step rolling method is used on the test set to recursively output the predicted values ​​of the three feature components and their confidence intervals for each future window, and the predicted values ​​of each window are connected to form a prediction trend curve. The early warning judgment stage compares the predicted value with the boundary values ​​of its historical normal fluctuation range. When the predicted value deviates from the historical mean by more than twice the historical standard deviation, an abnormality early warning signal is triggered. After each round of prediction, the root mean square error of the prediction is calculated using the actual values ​​of the test set. When the error exceeds the tolerance limit, the model is automatically re-estimated to adapt to long-term changes in the data pattern.

[0047] The predicted trend curves, abnormal warning signals, and confidence intervals are organized by industry and time window to form a set of time evolution characteristic parameters. This set of parameters includes the future contribution trend direction, fluctuation range, and warning time nodes of each industry.

[0048] Step S6: Based on the time evolution characteristic parameter set and the spatial distribution data of the water-receiving area, construct a spatiotemporal evolution analysis model of the water-saving effect of the water-receiving area, and determine the spatiotemporal distribution map of the water-saving impact contribution value of the water-receiving area.

[0049] In one specific embodiment, the process of performing step S6 may specifically include the following steps: Acquire spatial distribution data of each area in the water-receiving zone, map each water user to the corresponding spatial grid cell based on the corresponding geographic coordinates, and establish a matrix of the attribution relationship between water users and spatial location; Using the time evolution feature parameter set as the time series dimension input and the attribution relationship matrix as the spatial dimension constraint, the spatiotemporal data fusion algorithm is used to sort and distribute the contribution of each water user to the spatial grid cell, and a spatiotemporal evolution analysis model of water-saving effectiveness in the water-receiving area is constructed. Using a spatiotemporal evolution analysis model, the water-saving impact contribution value of each spatial grid unit in different time windows is calculated, and a spatiotemporal distribution map of the water-saving impact contribution value of each region in the water-receiving area is output.

[0050] Specifically, spatial distribution data for each area within the water-receiving region is acquired, and each water user is mapped to a corresponding spatial grid cell based on their geographic coordinates, establishing a matrix relating water users to their spatial locations. The spatial distribution data includes the administrative boundaries of the water-receiving region, the distribution of major water systems, the routes of water diversion projects and pipelines, and the geographic coordinates of water intakes or water facilities for each water user. Data sources include land use maps from the natural resources department, water supply network maps from the water resources department, and business registration information for enterprises. The spatial grid cells are divided using a regular grid method, with grid side lengths determined based on the total area of ​​the water-receiving region and the density of the spatial distribution of water users, balancing the precision of spatial representation with the complexity of subsequent calculations. The specific determination rules are as follows: The target range for the number of water users within a single grid cell is preset to be 5 to 15. The total number of water users in the water-receiving area is counted. The average density area is obtained by dividing the total area of ​​the water-receiving area by the total number of water users. This average density area is then multiplied by the upper and lower limits of the target range and the square root is taken to obtain the range of grid side length values. Grid side lengths are selected within this range to ensure that the number of water users within a single grid cell falls within the target range. The mapping process iterates through the geographical coordinates of each water user, performs topological inclusion judgment on their relationship with the boundary range of the spatial grid cell, and determines the grid cell number to which each water user falls, forming a mapping relationship table between water users and grid cells. The attribution matrix uses the spatial grid cell number as the row index and the water user identifier code as the column index. Matrix elements are binary identifiers; when the geographical coordinates of a water user are located inside a spatial grid cell, the corresponding element is 1, otherwise it is 0. Each row vector completely describes the set of water users contained within that grid cell, and each column vector labels the grid cell to which the water user belongs spatially.

[0051] Using a time-evolution feature parameter set as the temporal dimension input and an attribution matrix as the spatial dimension constraint, the contribution ranking of each water user is distributed to spatial grid cells through a combination of matrix multiplication and weighted averaging, thus constructing a spatiotemporal evolution analysis model for water-saving effectiveness in the water-receiving area. The time-evolution feature parameter set provides the contribution ranking value and trend information of each water user within each time window, while the attribution matrix provides the attribution constraints between each water user and the spatial grid cells. The fusion operation uses water users as the linking bridge and executes window-by-window in a cyclical manner: for a single time window, the contribution ranking values ​​of all water users within that window are extracted from the time-evolution feature parameter set, forming a contribution vector of length equal to the total number of water users. This contribution vector is then multiplied by the attribution matrix, which serves as the spatial allocation weight matrix in the operation, spatially allocating the contribution ranking values ​​of each water user according to their respective grid cells. When multiple water users exist within a grid cell, the weighted average of their contribution ranking values ​​is taken as the comprehensive contribution value of that grid cell. The weighting coefficients are the proportion of each water user's water consumption to the total water consumption of that grid cell, and the water consumption data is retrieved from the water consumption structure distribution information of the water-receiving area. After weighted averaging, each grid cell obtains a comprehensive contribution value for each time window. The comprehensive contribution values ​​of all grid cells across all time windows constitute a two-dimensional spatiotemporal data matrix with grid cells as rows and time windows as columns. This matrix is ​​the core data structure of the spatiotemporal evolution analysis model of water-saving effectiveness in the water-receiving area.

[0052] The water-saving impact contribution value of each spatial grid unit within different time windows is calculated using a spatiotemporal evolution analysis model, and a spatiotemporal distribution map of the water-saving impact contribution value for each region of the water-receiving area is output. When calculating the water-saving impact contribution value, each grid unit in the spatiotemporal data matrix is ​​traversed, and the comprehensive contribution value sequence of that grid unit within each time window is extracted. This sequence is then normalized based on the water consumption scale of that grid unit. The normalization benchmark uses the average comprehensive contribution value of that grid unit across all time windows, converting the absolute contribution value into a relative contribution rate relative to its own historical average. This eliminates the incomparability of absolute contribution values ​​between different grid units due to differences in total water consumption, allowing comparisons between grid units of different scales at a unified scale. The normalized relative contribution rate is the water-saving impact contribution value of that grid unit. The water-saving impact contribution values ​​of all grid units within each time window are organized by spatial coordinates and timestamps, and visualized using a geographic information system. Color gradients represent the level of contribution values, generating a spatiotemporal distribution map of the water-saving impact contribution value for each region of the water-receiving area. The map allows for switching between time windows to view the spatial pattern evolution of water-saving contributions.

[0053] This spatiotemporal distribution map visually demonstrates the spatial clustering characteristics of water-saving contribution values. The contiguous distribution areas of grid cells with high contribution values ​​form water-saving hotspots, while the clustered areas of grid cells with low contribution values ​​form water-saving coldspots. Through the distribution map sequence of continuous time windows, the spatial migration paths and range expansion and contraction trends of hotspots and coldspots can be tracked, providing spatial positioning basis and temporal evolution reference for subsequent formulation of differentiated water resource allocation schemes for water-scarce areas with different causes.

[0054] Step S7: Based on the spatiotemporal distribution map of the water-saving impact contribution value, formulate a water resource allocation plan for the water-receiving area and generate a water-saving effectiveness evaluation report.

[0055] In one specific embodiment, the process of performing step S7 may specifically include the following steps: The spatiotemporal distribution map of the contribution value of water conservation impact is clustered and scanned using the local spatial autocorrelation algorithm. The hot and cold spot analysis statistics of the contribution value of each spatial unit are calculated. Spatial units with hot and cold spot analysis statistics below the preset cold point threshold are identified as cold point areas. They are then classified into three types according to their causes: insufficient water diversion dependence, low water use efficiency, and unbalanced industrial structure, generating water shortage cause labels. Based on the labels of causes of water shortage and combined with the hierarchical evaluation index system for water-saving effectiveness, a water resource allocation plan for the water-receiving area is formulated. The water resource allocation plan of the water-receiving area is converted into scenario parameters and substituted into the water-saving effect time series model to simulate the change trajectory of water-saving effect after implementation, and outputs the comparison curve of contribution value before and after optimization and the prediction timetable for the elimination of abnormal labels. By integrating water scarcity cause labels, water resource allocation plans for water-receiving areas, and contribution value comparison curves, a water-saving effectiveness evaluation report is generated.

[0056] Specifically, a local spatial autocorrelation algorithm is used to perform clustering scans on the spatiotemporal distribution map of water-saving impact contributions, calculating hot and cold spot statistics for the contribution values ​​of each spatial unit. This algorithm uses the local Moran's index as the statistic. A spatial weight matrix is ​​established based on a preset spatial adjacency relationship, centered on each spatial grid unit. The adjacency relationship adopts the first-order Rook adjacency definition, meaning that grid units sharing boundaries or vertices are neighbors. For each grid unit, the local Moran's index is calculated between its own contribution value and the weighted average of the contributions of its neighboring units. The formula for this index is the standardized deviation of the grid unit's contribution value multiplied by the weighted sum of the standardized deviations of the neighboring unit's contribution values. A positive local Moran's index indicates a positive spatial correlation between the contribution values ​​of the grid unit and its neighboring units, i.e., both are clustered at high or low values; a negative local Moran's index indicates a negative spatial correlation between the contribution values ​​of the grid unit and its neighboring units, i.e., high values ​​are surrounded by low values ​​or low values ​​are surrounded by high values. Statistical significance was determined using a permutation test. The contribution values ​​of all grid units in the region were randomly rearranged several times. After each rearrangement, the local Moran's index for each grid unit was recalculated, and a null distribution was constructed. The actual calculated local Moran's index was compared with the null distribution to obtain the significance level. Grid units with a significantly positive local Moran's index and a contribution value below a preset cold point threshold were identified as cold point areas. The preset cold point threshold was determined as follows: by default, the mean of the water-saving impact contribution values ​​of all grid units in the region minus one standard deviation was used as the cold point threshold. When the proportion of cold points corresponding to this threshold was below 20% or above 30%, the proportion of cold points was controlled within the range of 20% to 30% according to the water resource management objectives of the water-receiving area, and the corresponding contribution value quantile was used as the threshold.

[0057] Spatial units identified as cold-spot areas are categorized by cause into three types: insufficient dependence on diverted water, low water efficiency, and industrial structure imbalance, generating water shortage cause labels. The cause classification follows a composite discrimination rule, sequentially analyzing industry water use data within each cold-spot grid unit: The diverted water dependence index for each industry within the grid unit is retrieved. If the diverted water dependence is higher than the regional average but the contribution value remains at the cold-spot level, it indicates that the diverted water supply has not been fully converted into water-saving effects, and the grid unit is marked as insufficient diverted water dependence. The water efficiency deviation index for each industry within the grid unit is retrieved. If the water efficiency deviation exceeds a preset water efficiency deviation threshold, it indicates that the industry's water efficiency is significantly lower than the benchmark level, and the grid unit is marked as low water efficiency. The following steps are taken: First, water consumption structure data for each industry within the grid cell is retrieved. If a high water-consuming industry's water intake exceeds a preset industry proportion threshold and its water-saving contribution rate is lower than the regional average, it indicates that an unbalanced industrial structure leads to insufficient water-saving contributions. This grid cell is then marked as having an unbalanced industrial structure. The preset industry proportion threshold is determined by collecting the multi-year average of water consumption proportions for each industry (industry, agriculture, and service) from the water resources bulletin of the province or river basin where the water-receiving area is located. Industries whose water intake exceeds 30% of the total water intake of the entire region are identified as high water-consuming industries, and 30% is used as the default industry proportion threshold. For grid cells that simultaneously meet multiple causal conditions, a single causal label is assigned in the following priority order: insufficient dependence on water diversion takes precedence over low water efficiency, and low water efficiency takes precedence over unbalanced industrial structure. Each cold spot grid cell generates a water shortage causal label record, containing four fields: grid cell code, geographic coordinate range, cold spot statistical value, and causal type.

[0058] Based on the labels of water scarcity causes and in conjunction with the stratified evaluation index system for water conservation effectiveness, a water resource allocation plan for the water-receiving area was formulated. The formulation process used cold-spot grid units as the basic implementation unit, matching differentiated allocation strategies to different cause types. For grid units with insufficient reliance on diverted water, the allocation strategy focuses on optimizing the timing of diverted water allocation and the scheduling of water transmission networks, shifting the diverted water supply window forward to match peak water demand, and adding storage facilities to improve the utilization rate of diverted water. Correspondingly, the target value for the water diversion dependence of the industry in this region is adjusted in the stratified evaluation index system for water conservation effectiveness. For grid units with low water use efficiency, the allocation strategy focuses on mandatory water quota management and water-saving technological transformation. Water use permits are lowered based on the benchmark value for water use efficiency in this industry in the stratified evaluation index system for water conservation effectiveness, and phased targets for reducing water use efficiency deviations are set. The allocation strategy for grid units with imbalanced industrial structure focuses on controlling total water consumption and guiding water rights trading. This involves restricting new water rights for high water-consuming industries through a water rights trading platform, encouraging the transfer of water rights to low water-consuming, high-value-added industries, and simultaneously adjusting the water-saving potential coefficient of the relevant industries in the tiered evaluation index system for water-saving effectiveness. The allocation strategies for each grid unit are then compiled and integrated to form a water resource allocation plan covering all cold-spot areas and differentiating the causes of their formation.

[0059] The water resource allocation plan for the water-receiving area is converted into scenario parameters and substituted into a time series model of water-saving effectiveness to simulate the trajectory of water-saving effectiveness changes after implementation. The scenario parameterization process quantifies each measure in the allocation plan into time series parameter adjustment items that the model can recognize, including the correction value of exogenous intervention variables corresponding to the adjustment of water supply timing, the contribution trend correction coefficient corresponding to the adjustment of water use efficiency benchmark values, and the industry contribution weight redistribution factor corresponding to water rights trading guidance. These scenario parameters are matched one by one according to grid units and time windows and substituted into the trained water-saving effectiveness time series model. The model drives the model to output the predicted trend of water-saving effectiveness in each industry after the implementation of the allocation plan in a recursive manner. Connecting the predicted values ​​of each window forms a curve of contribution value changes after simulation implementation. Simultaneously with outputting the simulation results, the predicted anomaly label status within each time window is extracted and compared with the anomaly label status before the implementation of the allocation plan. The industries and time nodes of anomaly label elimination are statistically analyzed, generating a prediction timetable for anomaly label elimination. The contribution value change curves before and after the implementation of the allocation plan are superimposed and compared, outputting a curve comparing the contribution values ​​before and after optimization, quantitatively demonstrating the improvement magnitude and timing of the allocation plan's water-saving contribution to each cold region.

[0060] This report integrates water scarcity causal tags, water resource allocation schemes for the water-receiving area, and contribution value comparison curves to generate a water-saving effectiveness evaluation report. The report is output as an electronic document. The main body is organized into chapters based on cold-spot regions. Each chapter includes an explanation of the water scarcity causal tags for that region, a list of specific measures for the corresponding allocation scheme and their implementation priorities, a visual chart comparing the contribution value before and after optimization, and a textual explanation of the timeline for predicting the elimination of abnormal tags. The appendix includes a series of graphs depicting the spatiotemporal distribution of water-saving impact contributions across the entire water-receiving area, as well as a parameter estimation report for the water-saving effectiveness time series model.

[0061] It is understood that the implementing entity of this application can be a water-saving performance evaluation system for the water-receiving area of ​​a water diversion project, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiment uses a server as an example for illustration.

[0062] refer to Figure 2 This figure compares the water-saving effects before and after the implementation of the configuration plan. It shows three curves reflecting the changes in the overall contribution value of water-saving effectiveness: the black solid line represents the historical observation value during the baseline period, the red dashed line represents the predicted trend under the condition of no intervention, and the green solid line represents the water-saving effect after the implementation of the configuration plan. The comparison shows that the effectiveness curve gradually rises after the implementation of the plan, forming a difference with the predicted curve without intervention, intuitively demonstrating the improvement effect brought about by the configuration plan.

[0063] The above describes a method for evaluating the water-saving effectiveness of a water diversion project in the receiving area, as described in an embodiment of this application. The following describes a system for evaluating the water-saving effectiveness of a water diversion project in the receiving area, as described in an embodiment of this application. Please refer to [link / reference]. Figure 3 One embodiment of the water-saving performance evaluation system for water diversion projects in the receiving area of ​​this application includes: The water use structure analysis module is used to collect historical water use data and water diversion and supply records of various industries in the water-receiving area, and to construct water use structure distribution information of the water-receiving area. The substitution effect calculation module is used to establish a dynamic coupling model between the diverted water volume and local water resources based on the water use structure distribution information of the water receiving area, and to calculate the substitution effect coefficient of the diverted water on local water resources. The evaluation system construction module is used to construct a hierarchical evaluation index system for water-saving performance of multiple industries in the water-receiving area based on the substitution effect coefficient, combined with the water use efficiency benchmark value and water-saving potential coefficient of various industries in the water-receiving area. The anomaly attribution identification module is used to identify water-saving anomalies in various industries in the water-receiving area through a hierarchical evaluation index system for water-saving effectiveness, obtain water-saving anomaly labels, and combine water rights trading data to attribute water-saving contributions and obtain contribution ranking. The time-series evolution analysis module is used to construct a time-series model of water-saving effectiveness in the water-receiving area based on water-saving anomaly labels and contribution ranking, analyze the time evolution law of water-saving effectiveness in the water-receiving area, and form a set of time evolution feature parameters. The spatiotemporal evolution modeling module is used to construct a spatiotemporal evolution analysis model of the water-saving effect of the water-receiving area based on the set of time evolution characteristic parameters and the spatial distribution data of the water-receiving area, and to determine the spatiotemporal distribution map of the water-saving impact contribution value of the water-receiving area. The scheme report generation module is used to formulate water resource allocation schemes for water-receiving areas based on the spatiotemporal distribution map of water-saving impact contribution values ​​and generate water-saving effectiveness evaluation reports.

[0064] Through the collaborative efforts of the aforementioned components, this application constructs a complete processing flow from water use structure analysis to configuration scheme output. After the water use structure analysis module provides basic data support, the substitution effect calculation module models the time-segmented reduction responses of local water sources in various industries before and after the intervention of diverted water, refining the traditional, general water-saving evaluation to the substitution effect coefficient level for different industries and time windows, thus accurately identifying the true substitution contribution of diverted water. The evaluation system construction module combines this coefficient with the water use efficiency benchmark and water-saving potential coefficient of each industry to form a multi-dimensional, hierarchical evaluation benchmark, providing a comparable standard for subsequent anomaly identification. The anomaly attribution identification module further incorporates water rights trading data, clarifying and ranking the true water-saving contribution of each water user under the premise of water rights transfer, solving the previous problem of accurately attributing water-saving effectiveness across industries. The time-series evolution analysis module incorporates the dynamic changes in anomaly labels and contribution ranking into the time series model, extending water-saving effectiveness evaluation from static judgment to dynamic early warning through the introduction of exogenous variables and trend prediction, improving the foresight of water resource management. The spatiotemporal evolution modeling module utilizes a spatial attribution matrix to distribute temporal evolution characteristic parameters onto spatial grid cells, generating a spatiotemporal distribution map of water-saving impact contributions. This achieves a leap from industry attribution to spatial positioning, intuitively revealing the hot and cold distribution patterns of water-saving effectiveness. The solution report generation module, based on hot and cold spot clustering analysis and water scarcity cause labels in the spatiotemporal distribution map, formulates differentiated water resource allocation schemes, and substitutes them into a time-series model for scenario simulation verification, outputting a comparison curve of contribution values ​​before and after optimization and a timeline for anomaly elimination prediction.

[0065] above Figure 3 The water-saving performance evaluation system for the water-diversion project receiving area in this application embodiment is described in detail from the perspective of modular functional entities. The water-saving performance evaluation equipment for the water-diversion project receiving area in this application embodiment is described in detail from the perspective of hardware processing.

[0066] Figure 4This is a schematic diagram of the structure of a water-saving performance evaluation device for a water diversion project receiving area, as provided in an embodiment of this application. The water-saving performance evaluation device 300 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 310 (e.g., one or more processors) and a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) storing application programs 333 or data 332. The memory 320 and storage media 330 can be temporary or persistent storage. The program stored in the storage media 330 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the water-saving performance evaluation device 300. Furthermore, the processor 310 may be configured to communicate with the storage media 330 and execute a series of instruction operations in the storage media 330 on the water-saving performance evaluation device 300 to implement the steps of the aforementioned water-saving performance evaluation method for a water diversion project receiving area.

[0067] A water-saving performance evaluation device 300 for a water diversion project's receiving area may further include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 4 The structure of the water-saving effect evaluation device for the water-receiving area of ​​a water diversion project shown does not constitute a limitation on the water-saving effect evaluation device for the water-receiving area of ​​a water diversion project provided in this application. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0068] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for evaluating the water-saving effectiveness of a water diversion project in the receiving area, characterized in that, include: S1. Collect historical water usage data and water diversion and supply records of various industries in the water-receiving area to construct information on the distribution of water usage structure in the water-receiving area; S2. Based on the water use structure distribution information of the water receiving area, establish a dynamic coupling model between the diverted water volume and local water resources, and calculate the substitution effect coefficient of the diverted water on local water resources. S3. Based on the substitution effect coefficient, and combined with the water use efficiency benchmark and water-saving potential coefficient of various industries in the water-receiving area, construct a hierarchical evaluation index system for water-saving performance of multiple industries in the water-receiving area. S4. Through the water-saving effectiveness hierarchical evaluation index system, identify water-saving anomalies in various industries in the water-receiving area, obtain water-saving anomaly labels, and combine water rights trading data to attribute water-saving contributions and obtain contribution ranking. S5. Based on the water-saving anomaly labels and contribution ranking, construct a time series model of water-saving effectiveness in the water-receiving area, analyze the time evolution law of water-saving effectiveness in the water-receiving area, and form a set of time evolution characteristic parameters. S6. Based on the set of time evolution characteristic parameters and the spatial distribution data of the water-receiving area, construct a spatiotemporal evolution analysis model of the water-saving effect of the water-receiving area, and determine the spatiotemporal distribution map of the water-saving impact contribution value of the water-receiving area. S7. Based on the spatiotemporal distribution map of the water-saving impact contribution value, formulate a water resource allocation plan for the water-receiving area and generate a water-saving effectiveness evaluation report.

2. The method according to claim 1, characterized in that, S1 includes: Collect information on water usage time periods, water consumption, and water sources in industrial, agricultural, and domestic water-receiving areas, and integrate them to generate historical water usage data; Collect water supply records of water diversion projects in the water receiving area, including water diversion volume, water diversion period, and allocation information of the water receiving area; The historical water usage data and water supply records are preprocessed to remove abnormal data, and then aligned and integrated according to time series to construct water usage structure distribution information for the water-receiving area.

3. The method according to claim 1, characterized in that, S2 includes: Based on the water use structure distribution information of the water receiving area, the local water resource usage before and after the water diversion intervention is statistically analyzed by industry and time window, the local water source reduction of each industry in each time window is calculated, the magnitude characteristics and time lag characteristics of the reduction are extracted, and a set of local water source response change characteristics by industry is formed. Using the amount of water diverted as the independent variable and the amount of local water source reduction as the dependent variable, and utilizing the local water source response change feature set, a substitution relationship curve is fitted for each industry. The substitution relationship curve is dynamically calibrated according to time windows, and a dynamic coupling model of water diversion and local water resources is established. The substitution effect coefficient of each industry under each time window is output through the dynamic coupling model.

4. The method according to claim 1, characterized in that, S3 includes: Obtain water use efficiency benchmarks and water-saving potential coefficients for various industries in the water-receiving area, and multiply the substitution effect coefficients as weighting adjustment factors with the water use efficiency benchmarks to establish water-saving performance evaluation benchmarks for each industry. The water-saving performance evaluation benchmark is decomposed and extracted into three indicators: water use efficiency deviation, water-saving contribution compliance rate, and water diversion dependence, and integrated to generate a multi-dimensional water-saving performance evaluation index set; Based on the aforementioned multi-dimensional water-saving performance evaluation index set, stratification thresholds are set for each index, and water-saving performance is stratified for various industries in the water-receiving area, thus constructing a stratified evaluation index system for water-saving performance across multiple industries in the water-receiving area.

5. The method according to claim 1, characterized in that, S4 includes: The water-saving performance evaluation index system is used to evaluate the water-saving performance of various industries in the water-receiving area in real time. Industries with water-saving performance below the preset performance threshold are marked as water-saving abnormal industries. A water-saving abnormal label containing an industry identification code and an abnormal deviation level is generated for each water-saving abnormal industry. Obtain transaction data from the water rights trading platform, extract the water rights transfer volume and direction of water users in various industries, count the water rights transferred out as positive water-saving contribution, and the water rights transferred in as negative water-saving contribution, and calculate the water-saving contribution of each water user. The water-saving contribution is aggregated by industry, and a mapping relationship is established between the industry water-saving contribution and the water-saving anomaly label. The water-saving contribution attribution results of the water-receiving area are generated, and the contribution ranking of each water user to the overall water-saving effect of the water-receiving area is output.

6. The method according to claim 1, characterized in that, S5 includes: Extract the ranking change features and distribution concentration features of the contribution ranking according to the time window to form a time series feature vector of the water-saving contribution structure of the water-receiving area; The occurrence time and duration of the water-saving anomaly label are used as external intervention variables and introduced into the time series feature vector. A time series prediction algorithm with exogenous variables is used to construct a time series model of water-saving effectiveness in the water-receiving area. Using the water-saving effectiveness time series model, the water-saving effectiveness of the water-receiving area in each future time window is predicted and analyzed, and the predicted trend curves and abnormal early warning signals of water-saving effectiveness in various industries are obtained, which are then compiled into a set of time evolution characteristic parameters.

7. The method according to claim 1, characterized in that, S6 includes: Acquire spatial distribution data of each area in the water-receiving zone, map each water user to the corresponding spatial grid cell based on the corresponding geographic coordinates, and establish a matrix of the attribution relationship between water users and spatial location; Using the time evolution feature parameter set as the time-series dimension input and the attribution relationship matrix as the spatial dimension constraint, the spatiotemporal data fusion algorithm is used to sort and distribute the contribution of each water user to the spatial grid cell, and a spatiotemporal evolution analysis model of water-saving effectiveness in the water-receiving area is constructed. Using the spatiotemporal evolution analysis model, the water-saving impact contribution value of each spatial grid unit in different time windows is calculated, and the spatiotemporal distribution map of the water-saving impact contribution value of each region in the water-receiving area is output.

8. The method according to claim 1, characterized in that, S7 includes: The spatiotemporal distribution map of the water-saving impact contribution value is clustered using the local spatial autocorrelation algorithm. The hot and cold spot analysis statistics of the contribution value of each spatial unit are calculated. Spatial units whose hot and cold spot analysis statistics are lower than the preset cold spot threshold are identified as cold spot areas. They are classified into three types according to their causes: insufficient water diversion dependence, low water use efficiency, and unbalanced industrial structure, and water shortage cause labels are generated. Based on the aforementioned water shortage cause labels and the aforementioned water-saving effectiveness stratified evaluation index system, a water resource allocation plan for the water-receiving area is formulated. The water resource allocation scheme of the water-receiving area is converted into scenario parameters and substituted into the water-saving effect time series model to simulate the change trajectory of water-saving effect after implementation, and outputs the comparison curve of contribution value before and after optimization and the prediction timetable for the elimination of abnormal labels. By integrating the water shortage cause labels, the water resource allocation plan of the water-receiving area, and the contribution value comparison curve, a water-saving effectiveness evaluation report is generated.

9. A water-saving performance evaluation system for water-receiving areas of water diversion projects, used to implement the method as described in any one of claims 1-8, characterized in that, include: The water use structure analysis module is used to collect historical water use data and water diversion and supply records of various industries in the water-receiving area, and to construct water use structure distribution information of the water-receiving area. The substitution effect calculation module is used to establish a dynamic coupling model between the diverted water volume and local water resources based on the water use structure distribution information of the water receiving area, and to calculate the substitution effect coefficient of the diverted water on local water resources. The evaluation system construction module is used to construct a hierarchical evaluation index system for water-saving performance of multiple industries in the water-receiving area based on the substitution effect coefficient, combined with the water use efficiency benchmark value and water-saving potential coefficient of each industry in the water-receiving area. The anomaly attribution identification module is used to identify water-saving anomalies in various industries in the water-receiving area through the water-saving effectiveness hierarchical evaluation index system, obtain water-saving anomaly labels, and combine water rights trading data to attribute water-saving contributions and obtain contribution ranking. The time-series evolution analysis module is used to construct a time-series model of water-saving effectiveness in the water-receiving area based on the water-saving anomaly labels and contribution ranking, analyze the time evolution law of water-saving effectiveness in the water-receiving area, and form a set of time evolution feature parameters. The spatiotemporal evolution modeling module is used to construct a spatiotemporal evolution analysis model of the water-saving effect of the water-receiving area based on the set of time evolution characteristic parameters and the spatial distribution data of the water-receiving area, and to determine the spatiotemporal distribution map of the water-saving impact contribution value of the water-receiving area. The scheme report generation module is used to formulate a water resource allocation scheme for the water-receiving area based on the spatiotemporal distribution map of the water-saving impact contribution value, and generate a water-saving effectiveness evaluation report.

10. A device for evaluating the water-saving effectiveness of a water diversion project in the receiving area, characterized in that, The device includes: a memory and at least one processor, wherein the memory stores instructions; The processor invokes the instructions in the memory to cause the water-saving effectiveness evaluation device for the water-receiving area of ​​a water diversion project to execute the water-saving effectiveness evaluation method for the water-receiving area of ​​a water diversion project as described in any one of claims 1-8.

Images