Regional reclaimed water utilization scheduling method and system based on dynamics

Abstract

The invention relates to the technical field of water resource management and optimal scheduling, and discloses a dynamic-based regional reclaimed water utilization scheduling method and system.The method comprises the steps that natural water volume, water quality conditions and water resource supply and demand data are collected, and a water source data set is obtained; analyzing the water source data set to evaluate the current reuse water utilization situation, and determining the reuse water utilization potential in combination with a target file; integrating the reclaimed water utilization potential and the regional dynamic characteristic data into a comprehensive data set, and inputting the comprehensive data set into a dynamic model to predict a reclaimed water optimal configuration scheme; and triggering a corresponding water supply strategy according to the optimal configuration scheme in combination with the purpose of the reclaimed water. According to the application, efficient scheduling of reclaimed water is realized through the kinetic model, the water resource utilization rate is improved, and regional economic and ecological requirements are met.

Description

Technical Field

[0001] This invention relates to the field of water resource management and optimization scheduling technology, and more specifically, to a dynamic-based regional reclaimed water utilization scheduling method and system. Background Technology

[0002] With the acceleration of urbanization and the increasing severity of water scarcity, the utilization of reclaimed water has become an important way to alleviate water resource pressure and promote sustainable development. In regional water resource management, the scientific scheduling and optimal allocation of reclaimed water resources has become a research hotspot, especially given the ever-increasing demand for comprehensive water resource utilization in modern cities. However, existing reclaimed water scheduling methods still have significant shortcomings in terms of dynamic analysis capabilities, supply-demand matching accuracy, and temporal scheduling flexibility, making it difficult to meet the needs of efficient utilization in complex scenarios.

[0003] Specifically, existing technologies exhibit significant limitations when facing complex scenarios with substantial regional differences, diverse water quality requirements, and dynamically changing user demands. First, traditional methods lack in-depth analysis of the dynamic characteristics of reclaimed water supply and demand, making it difficult to accurately predict supply and demand balance at different time scales. Second, existing technologies are insufficient in simulating water quality evolution patterns, failing to fully consider the temporal and spatial changes in water quality indicators during reclaimed water transmission and distribution, leading to low water quality management efficiency. Furthermore, their ability to capture fluctuations in user demand is limited, especially when there are significant differences in demand among various user types such as industry, agriculture, and municipal miscellaneous uses, making accurate demand forecasting and scheduling optimization difficult.

[0004] Therefore, how to comprehensively consider the dynamic characteristics of reclaimed water supply and demand, water quality change patterns, and regional characteristics in the scheduling of reclaimed water resources to improve the utilization efficiency and management level of reclaimed water resources, thereby achieving more efficient and precise scheduling and optimized allocation of reclaimed water resources to meet the needs of modern cities for comprehensive utilization of reclaimed water resources, has become an urgent technical problem to be solved. Summary of the Invention

[0005] This invention provides a dynamic regional reclaimed water utilization scheduling method and system, which solves the technical problem that existing technologies struggle to comprehensively consider the dynamic characteristics of reclaimed water supply and demand, water quality change patterns, and regional characteristics in scheduling, thereby improving the utilization efficiency and management level of reclaimed water resources and achieving more efficient and precise reclaimed water scheduling and optimized allocation.

[0006] This invention provides a dynamic-based regional reclaimed water utilization scheduling method and system, including: Firstly, a dynamic-based regional reclaimed water utilization scheduling method includes: collecting natural water quantity, water quality status, and water resource supply and demand data to obtain a water source dataset; analyzing the water source dataset to assess the current status of reclaimed water utilization and determining the reclaimed water utilization potential in conjunction with target files; integrating the reclaimed water utilization potential with regional dynamic characteristic data into a comprehensive dataset, and inputting the comprehensive dataset into a dynamic model to predict the optimal reclaimed water allocation scheme; and triggering the corresponding water supply strategy based on the optimal allocation scheme and the intended use of reclaimed water.

[0007] Furthermore, data on natural water quantity, water quality, and water resource supply and demand are collected to obtain a water source dataset, including: adjusting various parameters of the monitoring equipment according to environmental conditions and monitoring objectives; obtaining real-time monitoring data based on the adjusted monitoring equipment; filtering the real-time monitoring data to remove outliers and performing data calibration; simultaneously synchronizing various types of monitoring data to ensure uniform alignment of all types of data on the time axis; and obtaining a water source dataset based on the real-time monitoring data.

[0008] Furthermore, the analysis of the water source dataset assesses the current status of reclaimed water utilization and determines the potential for reclaimed water utilization in conjunction with the target document. This includes: extracting information on reclaimed water production capacity and demand distribution through multi-dimensional analysis based on the water source dataset; constructing an evaluation index system in conjunction with local reclaimed water utilization policies and target requirements; comprehensively scoring the current status of reclaimed water utilization based on the evaluation index system, and determining the potential for reclaimed water utilization in conjunction with regional economic development needs and ecological environment goals.

[0009] Furthermore, based on the aforementioned water source dataset, information on reclaimed water production capacity and demand distribution is extracted through multi-dimensional analysis. This includes: extracting key information from the dataset to obtain core data; performing trend analysis based on the core data and using historical data to predict future reclaimed water production capacity and demand distribution, obtaining prediction results; evaluating the current regional reclaimed water treatment capacity, water quality compliance, utilization rate, and distribution characteristics of the facilities through data mining and statistical analysis based on the prediction results and predefined reclaimed water utilization potential standards in the database; collecting and interpreting local reclaimed water utilization standard documents, planning objectives, and technical standards, combined with regional economic development plans, population growth forecasts, and environmental protection requirements; and then comprehensively analyzing the matching degree between reclaimed water utilization guidelines and actual needs, identifying constraints and development opportunities for reclaimed water utilization by comparing and analyzing the gap between the current situation and the objectives, and finally determining the specific utilization potential of reclaimed water.

[0010] Furthermore, the reclaimed water utilization potential and regional dynamic characteristic data are integrated into a comprehensive dataset, and the comprehensive dataset is input into a dynamic model to predict the optimal allocation scheme of reclaimed water. This includes: obtaining real-time economic activity and ecological environment status data based on the regional characteristic data; classifying and encoding the reclaimed water utilization potential and real-time economic activity and ecological environment status data to obtain encoded data; integrating the encoded data to obtain a comprehensive dataset; and preprocessing the comprehensive dataset to obtain a preprocessed integrated set.

[0011] Furthermore, the dynamic model includes: an input layer that receives the preprocessed integrated set; a dynamic layer with three dynamic layers to simulate changes in reclaimed water supply and demand, water quality evolution, and fluctuations in user demand; and an output layer that outputs an optimized reclaimed water configuration scheme based on the simulation results. The preprocessed integrated set is input into the input layer, which, upon receiving it, converts it into a feature vector conforming to the input format of the dynamic layer and transmits it to the dynamic layer. The dynamic layer simulates changes in reclaimed water supply and demand, water quality evolution, and fluctuations in user demand according to the three-layer dynamic structure. The supply and demand dynamic layer calculates the reclaimed water supply and demand balance at different times, and the water quality evolution layer predicts the reclaimed water's future performance. The system analyzes water quality trends during the transmission and distribution process, and examines the changing patterns of demand among various user groups. The output layer, based on simulation results from the dynamic layer, comprehensively considers multiple constraints including supply-demand balance, water quality compliance, and user satisfaction. A multi-objective optimization algorithm is used to generate an optimized reclaimed water allocation scheme, which includes the reclaimed water allocation volume for each water supply area, water supply timing, water quality assurance measures, and emergency dispatch plans. A feedback mechanism compares and analyzes actual operational data with predicted results, dynamically adjusting dynamic model parameters and allocation strategies to ensure the optimized scheme adapts to real-time changes in regional reclaimed water utilization needs. Finally, a validated and optimized reclaimed water allocation scheme is output.

[0012] Furthermore, the three-layer dynamic structure includes: a supply and demand dynamic layer: based on dynamic theory, a dynamic feedback dynamic model of reclaimed water supply and demand is constructed. By setting key variables such as population growth rate, industrial structure adjustment coefficient, and the degree of promotion of water-saving technology, the changing trend of reclaimed water supply and demand balance under different time scales is simulated; a water quality evolution layer: using a hydrodynamic and water quality coupled dynamic model, the integrated water quality parameters are used as initial conditions. Combined with regional hydrogeological characteristics, pollutant degradation coefficients, and water body self-purification capacity, the evolution of water quality indicators (such as chemical oxygen demand, ammonia nitrogen, total phosphorus, etc.) of reclaimed water in the transmission and distribution process is simulated over time and space; and a user demand fluctuation layer: by collecting historical water use data of different user types (such as industrial users, agricultural users, and municipal miscellaneous users), time series analysis methods are used to identify the influencing factors of demand fluctuation cycle and seasonal changes, production plan adjustments, and meteorological conditions. A user demand prediction dynamic model is established to dynamically capture short-term fluctuations and long-term trends in demand.

[0013] Furthermore, the dynamic model also includes: initializing the dynamic model and measuring the difference between the dynamic model output and the actual requirements to guide the dynamic model parameters; optimizing the difference and iterating the dynamic model parameters through multiple rounds to dynamically adjust and gradually approach the optimal solution; simulating and training the dynamic model according to the updated dynamic model parameters to obtain training results; and performing verification tests according to the training results to obtain the tested dynamic model.

[0014] Furthermore, based on the optimized configuration scheme and the corresponding water supply strategy triggered by the reclaimed water's intended use, the process includes: analyzing the user types of reclaimed water in the optimized configuration scheme and the water demand and quality requirements of each user type, and dividing the reclaimed water supply area accordingly; determining the water supply priority based on the analysis results and water quality index data, and planning four water supply areas centered on the supply area, with different water supply levels: Level 1, Level 2, Level 3, and Level 4; and establishing a mapping table between reclaimed water uses and water supply strategies, where industrial users are given priority in water supply strategies, and agricultural irrigation users are given priority in water supply strategies. The system employs a time-sharing water supply strategy, a dynamic water level maintenance strategy for ecological landscape users, and sets corresponding trigger water supply strategies for four different water supply areas. It monitors the effluent quality of the reclaimed water plant and the pressure at the end of the pipe network in real time. When water quality indicators deviate from the user's required threshold, the deep treatment unit is automatically activated. When the pipe network pressure is lower than the set value, the frequency conversion regulation system of the booster pump station is triggered. Simultaneously, it guides users within the area according to the water supply strategy. Based on user feedback data and operational energy consumption data, the water supply strategy is dynamically optimized periodically. The parameter weights in the mapping table are updated monthly, and resources outside the area are coordinated to promptly supplement areas with insufficient water supply.

[0015] Secondly, a dynamic-based regional reclaimed water utilization scheduling system includes: a data acquisition module, which collects natural water quantity, water quality, and water resource supply and demand data to obtain a water source dataset; a potential assessment module, which analyzes the water source dataset to assess the current status of reclaimed water utilization and determines the reclaimed water utilization potential in conjunction with target files; an optimization configuration module, which integrates the reclaimed water utilization potential with regional dynamic characteristic data into a comprehensive dataset and inputs the comprehensive dataset into a dynamic model to predict the optimal reclaimed water configuration scheme; and a water supply strategy module, which triggers the corresponding water supply strategy based on the optimal configuration scheme and the intended use of reclaimed water.

[0016] The beneficial effects of this invention are as follows: By constructing a three-layer dynamic model, this invention can accurately simulate the dynamic characteristics of reclaimed water supply and demand changes, water quality evolution, and user demand fluctuations. Specifically, the supply and demand dynamic layer, based on key variables such as population growth rate and industrial structure adjustment coefficient, achieves accurate prediction of supply and demand balance at different time scales, improving the accuracy of reclaimed water supply and demand dynamic prediction. Furthermore, by employing a coupled hydrodynamic and water quality dynamic model, combined with regional hydrogeological characteristics, pollutant degradation coefficients, and water body self-purification capacity, it can accurately simulate the evolution of water quality indicators over time and space during the transmission and distribution of reclaimed water, achieving precise simulation of the entire process of water quality evolution. Finally, through time series analysis of the user demand fluctuation layer, it can identify the demand fluctuation cycles and influencing factors of different user types, establishing an accurate user demand prediction dynamic model. This significantly enhances the ability to capture and respond to user demand fluctuations, especially in complex scenarios where there are significant differences in demand among various user types, such as industrial, agricultural, and municipal miscellaneous uses. Furthermore, a multi-objective optimization algorithm was used to comprehensively consider multiple constraints such as supply and demand balance, water quality compliance, and user satisfaction. This generated a complete optimized configuration scheme including the allocation of water to each supply area, water supply timing, water quality assurance measures, and emergency dispatch plans, achieving optimized configuration under multi-objective constraints. By establishing a mapping table between reclaimed water uses and water supply strategies, the system automatically triggered differentiated water supply strategies such as priority water supply for industrial users, timed water supply for agricultural irrigation, and dynamic water level maintenance for ecological landscapes. Real-time monitoring and feedback adjustment ensured dynamic optimization of the water supply strategies, establishing an intelligent water supply strategy triggering mechanism. Then, through a feedback mechanism, actual operating data was compared and analyzed with predicted results, dynamically adjusting the dynamic model parameters and configuration strategies. This enabled the system to adapt to the real-time changing needs of regional reclaimed water utilization, significantly improving system stability and reliability, and enhancing the system's adaptive adjustment capabilities. Finally, through a four-level water supply area division and a cross-regional resource coordination mechanism, areas with insufficient water supply could be promptly supplemented, improving the coordination of water resource allocation between regions and the overall utilization efficiency, achieving cross-regional resource coordination optimization. This invention effectively solves the technical limitations of existing reclaimed water scheduling methods in complex scenarios, and achieves more efficient and accurate reclaimed water scheduling and optimized allocation. It provides scientific and reliable technical support for the comprehensive utilization of reclaimed water resources in modern cities, and has important practical value and promotion significance. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the dynamic-based regional reclaimed water utilization scheduling method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a dynamic-based regional reclaimed water utilization scheduling system module provided in an embodiment of the present invention. Detailed Implementation

[0018] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, some features described in the examples may be combined in other examples.

[0019] This invention provides a dynamic-based regional reclaimed water utilization scheduling mechanism that establishes a four-stage linkage mechanism of perception, assessment, modeling, and decision-making, forming a complete chain from raw data acquisition to final scheduling execution. First, basic information on regional hydrology, water quality, and water use behavior is acquired through multi-source heterogeneous data collection methods to construct a comprehensive water source dataset reflecting the regional water resource status. Second, based on this dataset, a gap analysis is conducted between reclaimed water production capacity and actual demand, integrating policy guidance and development planning to quantify the regional reclaimed water utilization potential. Third, the assessed reclaimed water utilization potential value is fused and encoded with dynamic characteristic parameters such as the intensity of regional economic activity and the state of the ecological environment, and input into a specially designed dynamic model to simulate the evolution path of reclaimed water supply and demand over a future period, outputting an optimized configuration scheme including allocation quantity, water supply rhythm, and safeguard measures. Finally, based on the water use characteristics and water quality requirements of different users, corresponding water supply control logic is automatically matched to achieve precise and differentiated water supply scheduling. The dynamic model employs a three-layer dynamic structure (supply and demand dynamic layer, water quality evolution layer, and user demand fluctuation layer) to simulate the dynamic behavior of the reclaimed water system. The specific details of each layer are as follows: (1) Supply and demand dynamic layer: Based on the theory of system dynamics, a dynamic feedback model of reclaimed water supply and demand is constructed to simulate the changes in supply and demand balance at different time scales. Its differential equation is expressed as: Where d represents the differential symbol, and S(t) is the amount of reclaimed water supplied at time t; Let t represent the user demand at time t; Ps is the parameter vector of the supply and demand dynamic layer, including population growth rate, industrial structure adjustment coefficient (dimensionless), and degree of promotion of water-saving technology (dimensionless), etc. The supply and demand change function reflects the nonlinear feedback relationship between supply and demand, and is affected by economic, policy and environmental factors. (2) Water quality evolution layer: The hydrodynamic-water quality coupling model is used to simulate the spatiotemporal evolution of water quality indicators during the transportation and distribution of reclaimed water. Its differential equation is expressed as: Where C(t) is the water quality concentration vector at time t, including chemical oxygen demand (COD), ammonia nitrogen (NH3-N), total phosphorus, etc. t represents the processing efficiency (dimensionless); Pq is the parameter vector of the water quality evolution layer, including pollutant degradation coefficient, water body self-purification capacity, regional hydrogeological characteristic parameters (such as flow velocity, pipe material coefficient), etc.; g is the water quality evolution function, describing the migration, transformation and decay of pollutants during the transportation process. (3) User demand fluctuation layer: Based on historical water use data, a user demand prediction model is established using time series analysis to capture short-term fluctuations and long-term trends in demand. Its differential equation is expressed as: ;in: The user demand at time t (unit: m³ / d). Pd is the price influencing factor (dimensionless) at time t; Pd is the parameter vector of the user demand fluctuation layer, including the demand elasticity coefficient. (dimensionless), seasonal fluctuation parameter σ (dimensionless), industrial production cycle coefficient, etc.; h is the demand fluctuation function, reflecting the influence of economic, climate and policy factors on user behavior. (4) Overall model coupling and output: The three-layer dynamic structure is coupled to each other through the data interface: the output of the user demand fluctuation layer drives the supply and demand dynamic layer, the allocation result of the supply and demand dynamic layer is used as the input of the water quality evolution layer, and the simulation result of the water quality evolution layer is used to correct the supply and demand allocation. The output layer is based on the simulation result of the dynamic layer and uses a multi-objective optimization algorithm (such as genetic algorithm or particle swarm optimization) to generate a reclaimed water optimization configuration scheme, including: the reclaimed water allocation of each water supply area; water supply timing arrangement (such as staggered peak water supply); water quality protection measures (such as the layout of chemical dosing points and the triggering of deep treatment units); emergency dispatch plan (such as the activation of backup water sources). (5) Feedback mechanism and parameter update: The actual operation data is compared with the prediction results through the feedback mechanism to dynamically adjust the model parameters. The parameter update adopts the gradient descent method: Where θk is the model parameter vector for the k-th iteration; θk+1 is the model parameter vector for the (k+1)-th iteration; L(θk) is the learning rate (dimensionless); L(θk) is the loss function (such as mean squared error or weighted bias). L(θk) is the gradient of the loss function.

[0020] Parameter calibration is a crucial step in ensuring the accuracy of the dynamic model. Specifically, it involves setting initial values ​​for parameters based on historical statistical data, expert experience, or literature. For example, the population growth rate *rp* is initialized to the average growth rate over the past five years; the industrial structure adjustment coefficient *αi* is based on regional economic yearbook data; the pollutant degradation coefficient *kd* is referenced to the measured attenuation rate of a typical reclaimed water system; and the water body's self-purification capacity *γ* is quantitatively set based on regional hydrogeological conditions (such as flow velocity and dissolved oxygen levels). The loss function is defined using mean squared error (MSE), mean absolute percentage error (MAPE), or Nash efficiency coefficient (NSE) to measure the difference between the model output and actual demand. Intelligent optimization algorithms such as gradient descent, genetic algorithms, or particle swarm optimization are used to iterate the model parameters multiple times to minimize the loss function. For example, the industrial structure adjustment coefficient *αi* in the supply-demand dynamic layer is adjusted using gradient descent; the COD degradation coefficient *kd* in the water quality evolution layer is recalibrated using a genetic algorithm; and the demand elasticity coefficient in the user demand fluctuation layer is corrected using a Bayesian update method. The model is trained using historical datasets (covering scenarios such as normal water supply, extreme weather, and emergencies) to gradually approximate the optimal solution, improving the model's generalization ability. The constructed kinetic model is then validated to evaluate its performance and reliability on unseen data: using historical or future data independent of the training data as the validation set ensures data representativeness and independence. Validation metrics include: Root Mean Square Error (RMSE); Nash Efficiency Coefficient (NSE); and Mean Absolute Percentage Error (MAPE). Validation criteria are set with thresholds (e.g., MAPE ≤ 10% or NSE ≥ 0.8). If the model metrics meet the thresholds, validation is passed; otherwise, the model returns to the parameter calibration phase for readjustment. The model's sensitivity to changes in key parameters (such as population growth rate rp and pollutant degradation coefficient kd) is tested to ensure model robustness. For example, Monte Carlo simulations are used to analyze the impact of parameter uncertainty on the output results.

[0021] The embodiments provided by the present invention disclose a dynamic-based regional reclaimed water utilization scheduling method and system, including: like Figure 1 As shown, the dynamic-based regional reclaimed water utilization scheduling method includes the following steps: Step 1: Collect data on natural water quantity, water quality, and water supply and demand to obtain a water source dataset; The collection of natural water quantity, water quality, and water resource supply and demand data refers to the acquisition of these data through automated monitoring equipment deployed at key nodes in the watershed. This data includes multi-dimensional raw observations such as surface runoff, groundwater level, concentrations of major pollutants (e.g., COD, ammonia nitrogen, total phosphorus), effluent indicators from wastewater treatment plants, and user water intake. These data cover time scales from hourly to annual levels and spatially encompass upstream water inflow areas, treatment facility distribution zones, and downstream reuse areas. To ensure data quality, the system supports remote calibration and can access auxiliary information sources such as weather forecasts and land use change data to enhance data integrity. Obtaining the water source dataset involves cleaning, normalizing, and time-series aligning the aforementioned raw data to form a structured database, which serves as the foundation for subsequent analysis. This dataset not only includes instantaneous observations but also retains historical trend sequences, facilitating long-term evolution analysis. For example, in a demonstration area in the middle and lower reaches of the Fen River, the accuracy of natural water volume estimation can be improved by integrating water surface area data retrieved from remote sensing with flow data measured by on-site sensors. Alternatively, mobile rapid detection devices can be introduced to supplement blind spots at fixed stations, enhancing spatial representativeness. As an optional implementation, drone inspections combined with spectral imaging technology can be used to acquire large-scale preliminary water quality data, which can then be verified by fixed stations, achieving a combined data collection mode of wide-area screening and focused confirmation.

[0022] Step 2: Analyze the water source dataset to assess the current status of reclaimed water utilization and determine the potential for reclaimed water utilization in conjunction with the target document; The analysis of water source datasets involves using statistical analysis, machine learning algorithms, or multi-indicator comprehensive evaluation methods to extract key features from massive amounts of data. This identifies issues such as whether current reclaimed water treatment capacity is saturated, the coverage rate of major reuse areas, and whether there are any substandard water discharges. Assessing the current status of reclaimed water utilization is a diagnostic conclusion based on the extracted results, typically expressed as a regional utilization percentage, facility load rate, and a distribution map of reuse destinations. This is combined with binding or guiding policy texts in target documents, such as local special guidelines for reclaimed water utilization and ecological protection red line control requirements. By comparing the current status assessment results with the target requirements item by item, the gap between the current level and the expected target is identified. Determining the reclaimed water utilization potential involves deriving the potential increase in reuse volume, expandable application scenarios, and key areas requiring priority transformation in the future. For example, if the current industrial reuse rate is 20%, while the planned target is 45%, the difference constitutes potential incremental space. Similarly, if an agricultural irrigation area still relies on groundwater irrigation and has stable reclaimed water sources nearby, the area is identified as a high-potential development zone for reclaimed water utilization. As an alternative, scenario simulation can be introduced to set different development paths (such as high economic growth and strong environmental constraints), and the maximum potential for the utilization of reclaimed water under the corresponding conditions can be calculated to provide a flexible reference for decision-making.

[0023] Step 3: Integrate the reclaimed water utilization potential and regional dynamic characteristic data into a comprehensive dataset, and input the comprehensive dataset into the dynamic model to predict the optimal configuration scheme for reclaimed water; The regional dynamic characteristic data mainly includes information reflecting the real-time state of the socio-economic and natural environment, such as the daily or recent industrial production index, agricultural cropping cycle, temperature and precipitation forecasts, urban population flow trends, and arrangements for major public events. This data is highly time-sensitive and directly affects short-term water demand. Integrating it into a comprehensive dataset involves spatiotemporal matching and format unification of the static reclaimed water utilization potential assessment results with dynamic characteristic variables to form a joint input matrix that can be used for dynamic model calculations. For example, the reclaimed water utilization potential can be divided into two parts: basic reclaimed water utilization potential and flexible reclaimed water utilization potential. The former represents a long-term stable adjustable water volume, while the latter fluctuates with dynamic factors. A dynamic model is a simulation framework based on differential equations or system dynamics principles, capable of characterizing the complex nonlinear interaction between the reclaimed water supply and demand sides. Its core lies in simulating the trajectory of supply and demand balance changes over time, rather than simply providing the optimal solution at a single moment. This dynamic model can not only predict resource allocation paths under normal conditions but also simulate emergency response effects under sudden events (such as rainstorms leading to deterioration of influent water quality or sudden water outages). The predicted optimized allocation scheme for reclaimed water outputs a multi-dimensional decision-making recommendation package, encompassing recommended water supply volumes, suggested start-up and shutdown times, recommended water quality levels, risk warning levels, and contingency plans for each water supply zone. For example, during the high-temperature summer months, the dynamic model can anticipate the rising demand for landscape water and recommend storing high-quality reclaimed water in advance for replenishing park lakes; while during the dry winter months, it recommends prioritizing continuous industrial production water use and reducing unnecessary municipal water consumption. As an optional implementation, this dynamic model can also be embedded in a digital twin platform for visualized simulations and interactive adjustments.

[0024] Step 4: Trigger the corresponding water supply strategy based on the optimized configuration scheme and the intended use of reclaimed water; The optimized configuration scheme translates the macroscopic recommendations output by the dynamic model into specific, executable operational instructions, such as adjusting the pump station's operating frequency, switching valve pathways, and activating deep treatment units. It emphasizes tailored control logic based on the technical requirements of different application scenarios, considering the intended use of reclaimed water. For example, industrial cooling water is sensitive to turbidity and hardness, requiring high water quality stability; agricultural irrigation focuses on pathogen removal rates and the risk of salt accumulation; and ecological water replenishment emphasizes maintaining dissolved oxygen levels and nutrient balance. Triggering the corresponding water supply strategy means the system automatically activates the appropriate control process based on a preset rule base. For instance, when an industrial park enters its peak production period and has sufficient water reserves, it automatically switches to a priority water supply strategy, increasing water pressure and frequency; for farmland irrigation areas, a time-sharing water supply strategy is implemented according to the crop growth cycle to avoid peak evaporation periods; and for urban artificial lakes and other ecological landscape water bodies, a dynamic water level maintenance strategy is adopted, automatically replenishing an appropriate amount of reclaimed water based on evaporation and seepage rates. Furthermore, the system supports feedback closed-loop regulation—by monitoring the water quality and pressure at the end of the pipe network online, corrective actions are initiated once deviations from thresholds are detected, such as automatically adding disinfectant or calling up backup pump sets. In addition, the effectiveness of the strategy can be periodically evaluated based on user feedback and energy consumption statistics, and the weight parameters in the strategy mapping table can be dynamically updated to achieve adaptive optimization. As an extended implementation method, this strategy library can also be connected to a higher-level smart water management platform to receive cross-regional coordination instructions and participate in the joint regulation and supply of water resources on a larger scale.

[0025] Through the aforementioned steps, this application achieves a shift from passive response to proactive prediction, and from extensive management to refined control. By introducing a dynamic modeling mechanism, the system can capture the dynamic coupling relationships between various elements within the reclaimed water system, enhancing its ability to predict future trends. Integrating multi-source data with policy objectives strengthens the scientific and compliant nature of resource allocation. A differentiated strategy system oriented towards specific uses improves scheduling flexibility and user satisfaction. In particular, the entire method forms a complete closed-loop control chain of perception-assessment-modeling-execution, effectively solving the problems of strong lag and poor adaptability in traditional scheduling models, and significantly improving the overall efficiency and safety of regional reclaimed water utilization.

[0026] Based on the above preferred embodiment, this embodiment further proposes: collecting natural water quantity, water quality status, and water resource supply and demand data to obtain a water source dataset, including: adjusting various parameters of the monitoring equipment according to environmental conditions and monitoring objectives; obtaining real-time monitoring data based on the adjusted monitoring equipment; filtering the real-time monitoring data to remove outliers and calibrating and synchronizing various types of monitoring data to ensure uniform alignment of various types of data on the time axis; and obtaining a water source dataset based on the real-time monitoring data. The adjustment of various parameters of the monitoring equipment involves adjusting the sensitivity and cleaning frequency of the turbidity meter according to changes in water turbidity, adjusting the sampling interval of the flow meter according to the flow velocity range, or calibrating the temperature compensation coefficient of the pH sensor according to seasonal temperature changes. The monitoring equipment refers to sensing devices used to perceive and collect physical quantities related to hydrology, water quality, and water use behavior, including but not limited to flow meters, level gauges, conductivity meters, pH sensors, dissolved oxygen meters, turbidity meters, online COD (chemical oxygen demand) analyzers, and ammonia nitrogen detectors. These devices are deployed at river cross-sections, groundwater wells, sewage treatment plant outlets, key nodes of reclaimed water transmission and distribution networks, and water intake points of key water users, forming a regional dynamic water resource sensing network. Adjusting various parameters refers to optimizing the sensor's range, sampling frequency, sensitivity threshold, signal gain, and zero-point drift compensation coefficient based on on-site environmental conditions (such as temperature, humidity, and water flow velocity), the characteristics of the measured medium (such as suspended solids content and corrosivity), and the accuracy requirements of the measurement target. For example, in high-turbidity river sections, the transmission power and signal filtering intensity of optical sensors should be appropriately increased; in low-flow-velocity areas, the trigger threshold for ultrasonic flowmeters needs to be lowered to avoid missed detections. Furthermore, a reasonable data upload cycle (such as uploading every 5 minutes) needs to be set to balance data continuity and communication load. Optionally, some monitoring devices support adaptive parameter adjustment functions, automatically optimizing sampling strategies based on historical data trends. For example, when a rainfall event is detected causing a sudden increase in runoff, the system can dynamically increase the sampling frequency of stormwater-related indicators to once per minute, achieving refined capture of key processes. Another variant is to use an edge computing module pre-installed with a parameter template library, which automatically switches to the optimal parameter combination according to different seasons or hydrological stages, thereby improving the stability of equipment operation.

[0027] Real-time monitoring data is acquired based on the adjusted monitoring equipment. This real-time monitoring data refers to the set of raw observations continuously collected and transmitted by the aforementioned parameter-optimized monitoring equipment, covering natural water quantities (such as river flow, precipitation, and evaporation), water quality conditions (such as pH, turbidity, conductivity, chemical oxygen demand (COD), BOD5 (the concentration of dissolved oxygen consumed by microorganisms decomposing organic matter after a water sample is cultured at 20°C for five days), total nitrogen, total phosphorus, and fecal coliform count), and water resource supply and demand information (such as industrial water intake, agricultural irrigation water consumption, municipal water supply, and user-reported water demand). This step connects the dispersed monitoring equipment to a regional data center or cloud platform via wired (such as RS485, fiber optic) or wireless communication methods (such as NB-IoT, LoRa, 4G / 5G) to form a unified data access channel. The data acquisition process follows a timestamp mechanism to ensure that each record is accompanied by a precise time stamp (accurate to the second or even millisecond level), providing a foundation for subsequent time series analysis. To ensure data continuity and integrity, the system can be configured with a dual-channel redundant acquisition architecture: when the main channel is working normally, it prioritizes the output results from high-precision sensors; once a device failure or signal interruption is detected, it automatically switches to the backup sensor or nearby site for interpolation and estimation data to maintain an uninterrupted data flow. Simultaneously, it supports a breakpoint resume mechanism to ensure no data loss even in unstable network environments.

[0028] Real-time monitoring data undergoes filtering to remove outliers and is calibrated and synchronized to ensure uniform alignment across different data types on the timeline. Filtering eliminates high-frequency noise caused by electromagnetic interference, transient sensor disturbances, or external shocks. Common filtering methods include moving average filtering, median filtering, Kalman filtering, and wavelet denoising. For example, for drastically fluctuating dissolved oxygen data, a five-point moving average can be used to smooth short-term jumps; for sudden spikes (such as voltage spikes caused by lightning strikes), median filtering is used to remove them. Outlier removal involves identifying and excluding data points that significantly deviate from the normal range. Common methods include the 3σ rule (data exceeding the mean ± 3 standard deviations is considered outliers), box plots (IQR criterion), or the Isolation Forest algorithm based on machine learning. For example, if a pH reading suddenly jumps from 7.2 to 13.5 within a certain period while surrounding stations show no similar changes, it is considered a sensor false alarm and corrected or marked as invalid. Calibration compensates for zero-point drift or sensitivity decay that may occur in sensors during long-term operation. It can be achieved through periodic manual calibration (e.g., calibrating pH probes with standard buffer solutions) or automated online calibration (e.g., using multi-parameter water quality analyzers equipped with integrated cleaning-calibration units). Cross-validation using data from upstream and downstream sites can also be performed, utilizing spatial consistency constraints for indirect calibration. Synchronizing various monitoring data refers to unifying data from different devices, types, and sampling frequencies to the same time reference. For example, a weather station reports rainfall hourly, while a water quality analyzer uploads a set of data every 10 minutes; these need to be aligned to a uniform time granularity (e.g., a complete record every 10 minutes) using linear interpolation or spline interpolation methods. Synchronization operations are typically performed at the data middleware layer, efficiently utilizing the built-in time alignment functions of time-series databases (e.g., InfluxDB, TDengine).

[0029] The system acquires a water source dataset based on real-time monitoring data. This dataset is a structured collection of heterogeneous data from multiple sources, integrated after parameter optimization, real-time acquisition, filtering calibration, and time alignment. It is typically stored in a relational or time-series database, and its fields include timestamps, geolocation codes, monitoring indicator types, values, units, and data quality identifiers. This dataset serves as the foundation for subsequent assessments of reclaimed water utilization potential and inputs to the scheduling dynamics model; its completeness, accuracy, and timeliness directly impact the reliability of the entire system's decision-making. The data organization can be modeled using a three-dimensional spatial-temporal-indicator structure, supporting rapid retrieval and multi-dimensional slice analysis. For example, it can extract the change curves of all water quality indicators for a specific month in a sub-basin, or statistically analyze the daily water consumption trend of an industrial park. As an optional implementation, the system can also introduce a data credibility scoring mechanism, assigning confidence weights to each data point (e.g., based on equipment health status, historical error rate, and neighborhood consistency scores) for weighted use in subsequent dynamics models, further enhancing its anti-interference capabilities. Based on the foregoing, this application achieves a high-quality data acquisition process for natural water quantity, water quality status, and water resource supply and demand information. By precisely controlling the parameters of the monitoring equipment, the measurement accuracy and environmental adaptability of the raw data were improved; the timeliness and continuity of information were ensured through a real-time data acquisition mechanism; and the consistency and reliability of the data were significantly enhanced through filtering, noise reduction, calibration, and time synchronization processing. The resulting water source dataset has the characteristics of high fidelity, strong synchronization, and easy integration, providing a solid data foundation for subsequent assessment of the current status of reclaimed water utilization, analysis of reclaimed water utilization potential, and prediction of dynamic models. It effectively solves the problems of data distortion, time misalignment, and noise interference in traditional monitoring methods, thereby supporting the scientific decision-making and precise response of the regional reclaimed water dispatching system.

[0030] Based on the preferred embodiment described above, this embodiment further proposes: analyzing the water source dataset to assess the current status of reclaimed water utilization and determining the potential for reclaimed water utilization in conjunction with target documents; wherein, this step involves systematically processing and analyzing the collected water source dataset to scientifically and quantitatively assess the realistic foundation and development space of regional reclaimed water utilization. Traditional methods rely heavily on experience or local indicators when judging the potential for reclaimed water utilization, lacking a unified assessment framework, resulting in highly subjective and incomparable results, making it difficult to support precise scheduling decisions. This implementation method achieves standardized and dynamic determination of the potential for reclaimed water utilization by constructing a data-driven + policy-guided dual-track mechanism. Based on the water source dataset, information on reclaimed water production capacity and demand distribution is extracted through multi-dimensional analysis; wherein, multi-dimensional analysis refers to cross-analyzing the water source dataset from multiple dimensions such as time, space, user type, and water quality level to comprehensively reveal the spatial pattern and temporal characteristics of reclaimed water supply capacity and water demand. Specifically, the analysis considers the time dimension to examine water volume fluctuations across different seasons, months, and even daily scales; the spatial dimension to identify the geographical match between production capacity and demand by combining administrative divisions, watershed units, or water supply network zones; the user type dimension to differentiate water usage intensity and frequency for different purposes such as industrial cooling, agricultural irrigation, municipal miscellaneous uses, and ecological water replenishment; and the water quality dimension to classify the applicable levels of usable water bodies based on reuse standards. The extraction of reclaimed water production capacity is based on a comprehensive calculation of parameters such as the actual treatment scale of wastewater treatment plants, operating load rates, effluent quality compliance rates, and existing network coverage, reflecting the maximum sustainable water production of current facilities. Demand distribution information is extrapolated using auxiliary data such as water use census data, industry water quotas, population density distribution, and land use types, and spatially mapped in conjunction with urban development plans. To enhance the adaptability of the dynamic model, principal component analysis (PCA) or clustering algorithms can be used to reduce the dimensionality of high-dimensional data and identify dominant influencing factors. For example, in cities with extremely scarce water resources, population density and industrial output may become key variables in demand forecasting; while in ecologically sensitive areas, the focus is more on river baseflow maintenance and wetland replenishment needs. Furthermore, evapotranspiration from remote sensing imagery can be used as an alternative estimation method for agricultural irrigation water demand, improving the timeliness and coverage of data acquisition. An evaluation index system should be constructed in conjunction with local reclaimed water utilization policies and objectives. This system connects objective data, aiming to transform the qualitative requirements in policy texts into calculable and comparable quantitative indicators. This system typically includes three levels: the objective level (i.e., a comprehensive score of reclaimed water utilization potential), the criteria level (such as resource endowment, infrastructure, management efficiency, and environmental constraints), and the indicator level (specific measurable parameters). Indicator selection follows the principles of scientific rigor, representativeness, availability, and guidance. For example, the reclaimed water utilization rate reflects the current system's operational efficiency; the completion rate of planning targets measures the progress of policy implementation; reclaimed water consumption per unit of GDP reflects the level of water-saving society construction; and the reclaimed water substitution rate in key industries highlights the direction of structural optimization.For ecological and environmental objectives, specific indicators such as ecological water use security rate and water body self-purification capacity recovery index can be set. Weight allocation can employ the Analytic Hierarchy Process (AHP) or the entropy weight method; the former is suitable for scenarios dominated by expert experience, while the latter is more suitable for objective weighting supported by big data. Simultaneously, differentiated weight combinations can be set according to the development stage of different regions—for example, emerging industrial parks should emphasize industrial reuse indicators, while older urban areas should emphasize pipeline coverage and public acceptance. This evaluation system has good scalability; in the future, it can be integrated with a carbon emission accounting module to add green and low-carbon indicators such as carbon emission reduction equivalent of reclaimed water utilization.

[0031] The current status of reclaimed water utilization is comprehensively scored based on an evaluation index system. Combined with regional economic development needs and ecological environment goals, the potential for reclaimed water utilization is determined. The comprehensive scoring formula is: Comprehensive Score = α × Production Capacity + β × Demand Distribution + γ × Policy Compliance, where α, β, and γ are the weighting coefficients of each indicator, and α + β + γ = 1. The comprehensive score is a weighted summation process after normalizing the indicators, outputting a dimensionless value between 0 and 1, representing the degree to which the current level of reclaimed water utilization in the region approaches the ideal state. The scoring process supports multi-scenario simulations, such as setting a baseline scenario (maintaining the status quo), an active promotion scenario (increasing investment), and a strict control scenario (strengthening regulations), to predict the evolution trend of reclaimed water utilization potential under different policy paths. The reclaimed water utilization potential is not a static value but a dynamic range; its upper limit is constrained by technical and economic feasibility, and its lower limit depends on the ecological protection red line. Therefore, in addition to the scoring, it is necessary to integrate regional medium- and long-term development plans, such as the reclaimed water utilization rate targets, the proportion of water consumption reduction per 10,000 yuan of industrial added value, and the requirements for ecological flow protection of rivers and lakes proposed in documents like the National Water Conservation Action Plan, the Outline for Ecological Civilization Construction, and the Urban Master Plan, to form a current situation-target gap analysis matrix. Based on this, the Delphi method or system dynamics model is used to identify and rank key constraints (such as lagging pipeline construction, imperfect pricing mechanisms, and insufficient user awareness) and development opportunities (such as centralized water supply in new industrial parks and the construction of smart water management platforms), ultimately outputting a graded and categorized map of reclaimed water utilization potential. For example, reclaimed water utilization potential can be divided into high, medium, and low levels, with priority development areas and areas to be temporarily deferred, providing a decision-making basis for subsequent optimal allocation. As an optional implementation method, fuzzy comprehensive evaluation methods can be introduced into the scoring dynamics model to handle uncertainties or linguistic descriptions (such as "good" or "basically satisfied") in some indicators, improving the robustness of the dynamics model. This application achieves a systematic and structured assessment of reclaimed water utilization potential. By employing multi-dimensional data analysis, the understanding of regional supply and demand patterns is deepened, avoiding the risk of misjudgment caused by single indicators. The construction of a policy-oriented evaluation indicator system enhances the transparency and credibility of the assessment process, addressing the issue of strong subjectivity inherent in traditional methods. The comprehensive scoring integrates the dual goals of economic development and ecological protection, ensuring that the direction of reclaimed water utilization aligns with both current conditions and long-term objectives. This implementation method provides a key input parameter for the dynamic model—reclaimed water utilization potential—supporting the dynamic simulation and optimization decisions of subsequent resource allocation, significantly improving the scientific rigor and foresight of regional reclaimed water scheduling.

[0032] Based on the above preferred embodiment, this embodiment further proposes: extracting information on reclaimed water production capacity and demand distribution through multi-dimensional analysis of the water source dataset, including: extracting key information from the dataset to obtain core data; performing trend analysis based on the core data and using historical data to predict future reclaimed water production capacity and demand distribution, obtaining prediction results; based on the prediction results, combined with the predefined reclaimed water utilization potential standards in the database, evaluating the current regional reclaimed water treatment capacity, water quality compliance, utilization rate, and distribution characteristics through data mining and statistical analysis methods; collecting and interpreting local reclaimed water utilization standard documents, planning objectives, and technical standards, combined with regional economic development planning, population growth forecasts, and environmental protection requirements; and then comprehensively analyzing the matching degree between reclaimed water utilization guidance and actual demand, identifying the constraints and development opportunities of reclaimed water utilization by comparing and analyzing the gap between the current situation and the objectives, and finally determining the specific utilization and potential of reclaimed water. The core data is obtained by extracting key information from a water source dataset. This dataset refers to a collection of natural water quantity, water quality, and water supply and demand information collected through monitoring equipment, encompassing multi-source heterogeneous data such as surface water, groundwater, wastewater treatment plant effluent, pipeline distribution data, and user water usage behavior. After filtering, calibration, and time-axis alignment, this dataset forms a structured or semi-structured database. Extracting key information involves using data cleaning, field filtering, and feature selection techniques to identify variables related to reclaimed water production capacity and demand from the raw data, such as wastewater treatment volume, influent COD concentration, average daily water supply pressure, and the proportion of industrial water use. For example, in the application scenario of typical cities in the middle and lower reaches of the Fen River, key indicators can be extracted, such as the design scale of wastewater treatment plants, actual operating load rate, environmental capacity of major receiving water bodies, and water quotas for key industries. To achieve efficient extraction, an ETL (Extract-Transform-Load) process or a data extraction module based on a rule engine can be used. Alternatively, feature importance ranking algorithms from machine learning (such as XGBoost FeatureImportance or PCA principal component analysis) can be introduced to automatically identify the most representative input variables. Furthermore, a dynamic threshold mechanism can be set to trigger an alert and prioritize inclusion in the core dataset when the change in a certain type of data exceeds a preset range. One optional implementation is to extract only continuously recorded, complete site data from the past five years, excluding abnormal time periods caused by equipment failures, thereby improving the reliability of subsequent analysis. Trend analysis is performed based on the core data, and historical data is used to predict future reclaimed water production capacity and demand distribution, yielding prediction results. Trend analysis refers to using time series modeling methods to identify the regular characteristics of reclaimed water-related parameters over time, including long-term growth trends, seasonal fluctuations, and periodic changes.Historical data typically covers at least three years of continuous observation records to ensure a sufficient sample base for training the dynamic model. Commonly used predictive dynamic models include ARIMA (Autoregressive Integral Moving Average), SARIMA (Seasonal ARIMA), exponential smoothing, and deep learning-based LSTM (Long Short-Term Memory) networks or Transformer architectures. In practical implementation, separate sub-dynamic models for predicting reclaimed water production capacity and demand distribution can be established. Production capacity prediction uses wastewater treatment plant capacity, equipment availability, and upgrade / renovation plans as inputs; demand distribution prediction incorporates factors such as municipal greening irrigation cycles, agricultural irrigation seasons, and industrial park production plans to construct a multivariate regression dynamic model. For example, in arid northern regions, summer ecological water replenishment demand increases significantly; forecast accuracy can be optimized by overlaying meteorological forecast data (such as precipitation probability and evaporation). An optional approach is to use ensemble learning methods to fuse the outputs of multiple basic dynamic models to improve overall prediction robustness. Prediction results should include confidence intervals and sensitivity analysis reports to facilitate decision-makers' assessment of uncertainty risks.

[0033] Based on the prediction results and combined with the predefined reclaimed water utilization potential standards in the database, data mining and statistical analysis methods are used to assess the current treatment capacity, water quality compliance, utilization rate, and distribution characteristics of reclaimed water treatment facilities in the region. The reclaimed water utilization potential standards are a set of quantitative evaluation benchmarks pre-defined in the database, typically formulated by local governments or industry authorities. These standards include water quality requirements for different uses (e.g., GB / T19923 "Water Quality Standard for Industrial Water Use in Urban Wastewater Reuse"), reclaimed water consumption intensity per unit of GDP, reasonable range of facility load rate, and service coverage radius. These standards can be used to classify areas into high, medium, and low reclaimed water utilization potential levels. Data mining and statistical analysis methods include cluster analysis (e.g., K-means for classifying facility efficiency levels), principal component analysis for dimensionality reduction, association rule mining (Apriori algorithm to discover the correlation between facility aging and effluent exceeding standards), and spatial heat map visualization. For example, a GIS platform can be used to spatially match the treatment capacity of each wastewater treatment plant with the distance and water usage characteristics of its surrounding potential users, identifying contradictory areas with strong capacity but insufficient coverage or high demand but insufficient supply. The assessment specifically includes: whether the treatment capacity meets long-term planning requirements, whether the effluent quality consistently meets reuse standards (e.g., COD ≤ 50 mg / L, ammonia nitrogen ≤ 10 mg / L), whether the actual utilization rate is below the 60% warning line, and whether there are issues such as urban-rural imbalance in facility layout. An alternative technical approach is to use a fuzzy comprehensive evaluation method to weight and score the above indicators, outputting a comprehensive reclaimed water utilization potential index map.

[0034] This study collects and interprets local standards, planning objectives, and technical standards for reclaimed water utilization, taking into account regional economic development plans, population growth forecasts, and environmental protection requirements. Standard documents include the "Water-Saving Society Construction Plan," the "Urban Wastewater Treatment and Resource Utilization Development Plan," and local reclaimed water-specific plans and water abstraction permit management regulations. Technical standards cover recommended reclaimed water treatment processes, disinfection method selection guidelines, and pipe material specifications. Regional economic development plans focus on industrial restructuring directions (such as the exit or upgrading of water-intensive industries) and the increased water demand resulting from major projects. Population growth forecasts are derived from population census trend projection data released by the National Bureau of Statistics, used to estimate the increase in water consumption for domestic miscellaneous uses and public facilities. Environmental protection requirements are reflected in constraints such as water function zoning restrictions, cross-sectional water quality assessment targets, and ecological protection red line control. During implementation, Natural Language Processing (NLP) technology can be used to extract keywords and perform semantic analysis on policy texts to build a structured knowledge base, assisting in determining whether policy guidance encourages the use of reclaimed water for industrial cooling or river replenishment. For example, if a region explicitly requires newly built industrial parks to be equipped with reclaimed water pipeline networks, it indicates that the region has strong policy incentives. An alternative is to establish a policy influence factor matrix, transforming various planning elements into quantifiable adjustment coefficients, and participating in the dynamic model calculation of reclaimed water utilization potential.

[0035] A comprehensive analysis of the alignment between reclaimed water utilization guidelines and actual needs is then conducted. By comparing the current situation with the goals, the constraints and development opportunities for reclaimed water utilization are identified, ultimately determining the specific utilization potential of reclaimed water. Utilization guidelines refer to the development direction of reclaimed water proposed in policies or plans, such as prioritizing industrial use, followed by landscaping, or promoting reclaimed water for toilet flushing across the entire region. Actual needs are based on real application scenario preferences derived from user surveys and industry water usage statistics. Mismatches may exist between the two; for example, policies may encourage agricultural irrigation, but farmers may be unwilling to accept it due to concerns about water quality safety. Comparing the current situation with the goals can be accomplished by setting key performance indicators (KPIs), such as current reclaimed water utilization rate vs. planned target value, existing pipeline length vs. coverage demand length, annual fresh water replacement volume vs. target replacement volume, etc. A larger gap indicates higher potential for reclaimed water utilization, but it may also mean greater implementation difficulties. Constraints may include insufficient fiscal investment, lack of price incentive mechanisms, difficulties in cross-departmental coordination, and low public awareness; development opportunities may arise from the launch of major ecological projects, the promotion of water conservation and emission reduction through carbon peaking initiatives, and the accelerated construction of smart water management platforms. This application systematically extracts the production capacity and demand characteristics of reclaimed water from massive heterogeneous data, enhances the scientific rigor and foresight of reclaimed water utilization potential assessment through trend prediction and standard comparison mechanisms, and integrates policy interpretation and gap diagnosis to accurately identify bottlenecks and development windows. By introducing a multi-level data analysis chain, it solves the problems of information fragmentation, inaccurate predictions, and unclear identification of constraint factors in traditional assessments, thus providing more operational decision support for regional reclaimed water scheduling and improving resource allocation efficiency and policy responsiveness.

[0036] Based on the preferred embodiment described above, this embodiment further proposes: integrating reclaimed water utilization potential and regional dynamic characteristic data into a comprehensive dataset, and inputting the comprehensive dataset into a dynamic model to predict optimal reclaimed water allocation schemes. This includes: acquiring real-time economic activity and ecological environment status data based on regional characteristic data; classifying and encoding the reclaimed water utilization potential and real-time economic activity and ecological environment status data to obtain coded data; integrating the coded data to obtain a comprehensive dataset; and preprocessing the comprehensive dataset to obtain a preprocessed integrated set. The acquisition of real-time economic activity and ecological environment status data based on regional characteristic data; wherein, regional characteristic data refers to a multi-source heterogeneous information set reflecting the socio-economic operation, population distribution, industrial layout, climate conditions, hydrogeological structure, and ecosystem health level within a specific geographical area. This type of data can be obtained through government-published statistical yearbooks, remote sensing monitoring platforms, environmental quality bulletins, mobile signaling data, power load curves, etc., and has the characteristics of high spatial resolution and fine temporal granularity. Real-time economic activity data refers to information reflecting the intensity of regional production, consumption, and service activities within the current calendar day, such as industrial electricity load change curves, traffic flow indices, pedestrian density in commercial areas, and the operating rate of key enterprises. This data can be used to characterize the fluctuations in actual water demand across different industries. Such data typically originates from smart meter systems, city brain platforms, or third-party big data service providers, with sampling frequencies ranging from minutes to hours. As an optional implementation, when an industrial park is shut down during holidays, its corresponding industrial water demand weight can be dynamically adjusted downwards within the day's economic activity, thereby enhancing the dynamic model's ability to respond to real-world scenarios. Ecological and environmental status data includes, but is not limited to, dissolved oxygen concentration in surface water, air temperature, precipitation intensity, soil moisture, vegetation cover index (NDVI), and blue-green spatial distribution, used to assess the carrying capacity of natural systems for reclaimed water reuse and environmental risks. For example, under sustained high temperatures and drought conditions, the demand for ecological water replenishment increases, while the self-purification capacity of water bodies decreases, requiring corresponding adjustments to water quality protection measures and water supply priorities. As another optional implementation, drones equipped with multispectral sensors can be used to regularly inspect river water quality, supplementing data blind spots at fixed monitoring stations and improving the accuracy of ecological and environmental status perception. The two types of data mentioned above, together with the potential for reclaimed water utilization, constitute dynamic input elements. By integrating real-time economic and social behavior with natural ecological responses, the spatiotemporal sensitivity and situational adaptability of the scheduling system are enhanced.

[0037] Classification and coding are performed on reclaimed water utilization potential and real-time economic activities and ecological environment status data to obtain coded data. Reclaimed water utilization potential is a quantitative expression of the regional reclaimed water exploitability, applicable scenarios, and spatial distribution characteristics derived from water source dataset analysis. It includes multiple dimensions such as treatment capacity redundancy, water quality compliance level, pipeline accessibility, and user matching degree. This reclaimed water utilization potential value can be represented as a standardized score (e.g., a 0–1 interval) or as a graded label (e.g., high / medium / low reclaimed water utilization potential area). Classification coding refers to the process of converting raw data into structured digital identifiers according to unified rules, aiming to eliminate data type differences and achieve cross-domain information alignment. Specifically, discretization methods are used to divide continuous variables into several intervals and assign category codes (e.g., K-means clustering or equal-frequency binning), while qualitative descriptive fields (e.g., light industry-dominated urban areas) are mapped to integer codes or one-hot vectors. For multidimensional reclaimed water utilization potential indicators, a weighted comprehensive index can be constructed and then graded and coded. As a variant implementation, a semantic coding mechanism can also be introduced. Natural language processing technology is used to transform unstructured expressions in policy documents (such as encouragement for priority use in municipal miscellaneous purposes) into computable constraint codes, which are then embedded into subsequent dynamic model calculations. Furthermore, the coding system supports extensibility, allowing new data types (such as public health emergencies) to be quickly integrated into the system through preset coding rules. This step achieves the transformation of heterogeneous data from its raw form to a dynamic model-friendly format, providing a syntactic and semantic consistency foundation for subsequent integration.

[0038] Encoded data is integrated to obtain a comprehensive dataset. This integration is a data fusion operation performed under a unified spatiotemporal benchmark, ensuring that all fields are aligned on the same geographic grid unit (e.g., a 1km × 1km raster) and time step (e.g., daily or hourly). The integration process uses a relational database or a spatio-temporal cube structure to organize the data, establishing a wide table structure with region ID + timestamp as the primary key. Each row represents a snapshot of a region's state at a specific moment. Data missing information handling strategies are considered during integration, such as using linear interpolation to fill short-term gaps, using the mean of similar regions to fill spatial gaps, or marking data as unknown categories to retain uncertainty. For conflicting data sources (e.g., differences in water quotas issued by different departments), priority rules or confidence weighting mechanisms are set for coordination. As an optional implementation, topological structure encoding required for Graph Neural Networks (GNNs) can be introduced, treating each region as a node and the water distribution paths between regions as edges, forming a graph-structured dataset to support more complex network modeling needs. This integration process breaks down the information barriers between static assessment of reclaimed water utilization potential and dynamic environmental perception, forming a high-dimensional input matrix with spatiotemporal correlation. The comprehensive dataset undergoes preprocessing to obtain a preprocessed integrated set. This preprocessing encompasses a series of data cleaning, normalization, dimensionality reduction, and feature engineering operations aimed at improving data quality and modeling efficiency. Key preprocessing methods include: outlier detection and removal (e.g., Z-score method or isolated forest algorithm), numerical normalization (Min-MaxScaling or Z-Score standardization) to eliminate dimensionality effects, and principal component analysis (PCA) or autoencoder for dimensionality compression to reduce noise interference from redundant features. Specifically, for time-series characteristics, a sliding window technique is introduced to construct historical state sequences as input features, enhancing the dynamic model's ability to remember trends. Simultaneously, derived variables such as the rate of change of the average rainfall over the previous three days and the year-on-year increase in industrial activity this week are added to enrich the contextual information. As an alternative implementation, the preprocessing process can be deployed on edge computing nodes, completing lightweight processing locally at the data acquisition end, reducing the load on the central server and improving response speed. This application realizes a technical approach that deeply integrates static reclaimed water utilization potential with dynamic regional operational status. By introducing real-time economic activity and ecological environment status data, the input of the dynamic model not only covers the long-term planned reclaimed water utilization potential but also reflects short-term disturbance factors. By classifying, coding, and systematically integrating multi-source heterogeneous data, the problem of different types of data being difficult to directly participate in modeling is solved. Furthermore, through a standardized preprocessing process, a high-quality, structured, and spatiotemporally aligned preprocessed integrated set is output, significantly improving the usability of input data and the prediction accuracy of the dynamic model.This provides a solid data foundation for subsequent dynamic model simulations of supply and demand evolution and the output of optimized allocation schemes, effectively supporting the scientific and timely nature of regional reclaimed water scheduling decisions.

[0039] Based on the preferred embodiment described above, this embodiment further proposes that the dynamic model includes: an input layer: receiving a pre-processed integrated set; a dynamic layer: setting three dynamic structures to simulate changes in reclaimed water supply and demand, water quality evolution, and fluctuations in user demand; wherein the three dynamic structures include a supply and demand dynamic layer, a water quality evolution layer, and a user demand fluctuation layer; and an output layer: outputting an optimized reclaimed water allocation scheme based on the simulation results of the dynamic layer; wherein the optimized reclaimed water allocation scheme includes the reclaimed water allocation volume, water supply timing arrangement, water quality protection measures, and emergency dispatch plan for each water supply area; and comparing the actual operating data with the prediction results through a feedback mechanism to dynamically adjust the dynamic model parameters and configuration strategies to adapt to real-time changing demands. Specifically, the first layer of the dynamic layer, simulating changes in reclaimed water supply and demand, can be represented as: ;in, For the change, Let be the amount of reclaimed water supplied at time t. Let be the demand at time t. The second layer, simulating water quality evolution, can be represented as: (The function represents the change in supply and demand.) ;in, This represents the water concentration vector at time t+1. Let be the water concentration vector at time t. Let be the processing efficiency at time t, and g be the water quality evolution function; The third layer, simulating fluctuations in user demand, can be represented as: ;in, This represents the user demand at time t+1. Let t represent the user demand at time t. Let t be the price influencing factor at time t, and h be the demand fluctuation function.

[0040] Furthermore, the system incorporates dynamic adjustments through a feedback mechanism, allowing the model to gradually approach the optimal solution. After model initialization, the system measures the difference between the model output and the actual requirements, and iterates the model parameters multiple times using an optimization algorithm. For example, assuming the initial parameters are θ, the system updates the parameters using gradient descent, with the following update formula: ;in, This is the model parameter vector for the k-th iteration; This is the model parameter vector for the (k+1)th iteration; L is the learning rate (dimensionless); L() is the loss function (such as mean squared error or weighted bias). This is the gradient of the loss function.

[0041] The supply-demand dynamic layer, water quality evolution layer, and user demand fluctuation layer do not operate in isolation but are coupled through a data interface: the output of the user demand fluctuation layer serves as the input of the supply-demand dynamic layer, driving the supply-demand balance simulation; the flow and path information allocated by the supply-demand dynamic layer serves as the initial and boundary conditions for the water quality evolution layer, driving the water quality simulation; the results of the water quality simulation, in turn, may constrain or correct the supply-demand allocation scheme. These three elements form a closed-loop feedback, collaboratively evolving dynamic simulation core. This dynamic model aims to address the shortcomings of traditional water resource scheduling dynamic models in modeling multi-dimensional dynamic characteristics. In existing technologies, most dynamic models employ static or simplified dynamic structures, making it difficult to comprehensively depict the complex coupling relationship between supply, quality, and demand in the reclaimed water system. Especially in real-world scenarios with significant differences in regional water resource conditions, diverse user types, strict water quality standards, and frequent external environmental disturbances, single-structure dynamic models often lead to large prediction biases, response lags, and poor regulation flexibility. Therefore, an integrated dynamic model architecture with multi-level dynamic simulation capabilities is urgently needed to improve the accurate representation of the entire chain behavior of the reclaimed water system.

[0042] The input layer receives a preprocessed aggregate set. As the data entry module for the dynamic model, the input layer receives this preprocessed aggregate set generated by the front-end processing flow. This set includes classified and coded data on reclaimed water utilization potential, real-time economic activity data, and ecological environment status data. These data undergo standardization, normalization, and time alignment to ensure that information from different sources and with different dimensions can participate in subsequent modeling at a unified scale. The input layer typically consists of data interface components, supporting various communication protocols (such as MQt, HTTP, and OPCUA) to access online monitoring systems, SCADA platforms, or database servers. Optionally, the input layer can also integrate edge computing units to perform preliminary data verification and cache management locally, improving system response speed. For example, in a typical urban application, the input layer receives updated data packets every 15 minutes, covering key indicators such as industrial water load index, agricultural irrigation meteorological factors, municipal reuse pressure values, and COD concentration in wastewater treatment plant effluent.

[0043] The dynamic layer is designed with a three-layer dynamic structure to simulate changes in reclaimed water supply and demand, water quality evolution, and fluctuations in user demand. This three-layer dynamic structure includes a supply and demand dynamic layer, a water quality evolution layer, and a user demand fluctuation layer. The dynamic layer is the core computational unit of the entire dynamic model, employing a layered decoupling design to decompose the complex system behavior into three relatively independent but interconnected sub-processes for modeling. The first layer, the supply and demand dynamic layer, characterizes the equilibrium evolution process between the regional reclaimed water supply capacity and the water demand of various users over time. This layer constructs a nonlinear feedback loop based on system dynamics theory, considering the impact of factors such as population growth rate, industrial structure changes, and the degree of implementation of water conservation policies on the supply and demand relationship, and describes its dynamic trajectory through a system of differential equations. The second layer, the water quality evolution layer, focuses on the physicochemical changes of reclaimed water in the transmission network. This layer introduces a hydrodynamic-water quality coupled dynamic model, combining parameters such as water flow velocity, pipe material, residence time, temperature, and pollutant degradation coefficients to simulate the attenuation or accumulation trends of key water quality indicators (such as COD, ammonia nitrogen NH3-N, and total phosphorus TP) along the spatial path. The third layer is the user demand fluctuation layer. It establishes a demand forecasting sub-dynamic model based on time series analysis to address the differences in water usage patterns among different types of users, such as industrial, agricultural, and ecological landscape users. This model identifies characteristics such as periodic fluctuations (e.g., seasonal agricultural irrigation peaks) and sudden events (e.g., water outages due to large-scale industrial maintenance), enabling the coordinated capture of short-term and long-term demand trends. The three-layer structure can operate in parallel or, as needed, establish information exchange channels to form a cross-layer feedback mechanism. For example, when the water quality evolution layer detects an increased risk of total phosphorus exceeding the standard at the end of a certain section of the pipeline network, it can send an early warning signal to the supply and demand dynamic layer, triggering a reduction in water supply or a switching of water sources. The output layer, based on the simulation results of the dynamic layer, outputs an optimized reclaimed water allocation scheme. This scheme includes the reclaimed water allocation volume for each water supply area, water supply timing arrangements, water quality assurance measures, and emergency dispatch plans. The output layer is responsible for converting the simulation results of the dynamic layer into an executable set of dispatch instructions. The generated reclaimed water optimization scheme not only includes basic water allocation decisions (i.e., the specific amount of water each water supply area should receive), but also detailed water supply schedules (such as time-sharing peak-shaving water supply plans), water quality assurance measures (such as the deployment of intermediate chemical dosing points and the conditions for activating secondary disinfection), and emergency plans (such as the order of activation of backup water sources under extreme weather conditions). The scheme is output in a structured data format, compatible with the scheduling engines of mainstream water management systems (such as GIS platforms and DMA zone management systems). For example, during periods of high temperature and drought in summer, the output layer may recommend prioritizing the supply of cooling water to industrial parks, while reducing the frequency of ecological water replenishment and deploying mobile deep treatment equipment in advance to the entrance of landscape lakes prone to algae growth.

[0044] The feedback mechanism compares actual operating data with predicted results, dynamically adjusting the parameters and configuration strategies of the dynamic model to adapt to real-time changing demands. This feedback mechanism constitutes a key component of the closed-loop control of the dynamic model. It periodically collects actual operating data from the reclaimed water system (such as actual water intake, measured water quality values, user complaint records, and energy consumption statistics) and compares it with the predicted output of the dynamic model from the previous period, calculating error indicators (such as root mean square error (RMSE) and Nash efficiency coefficient (NSE). If the deviation exceeds a preset threshold, an adaptive parameter adjustment algorithm (such as genetic algorithm (GA), particle swarm optimization (PSO), or Bayesian update method) is activated to correct sensitive parameters within the dynamic model (such as demand elasticity coefficient, pollutant decay rate, and pipeline leakage rate), thereby improving subsequent prediction accuracy. Furthermore, the feedback mechanism supports strategy-level adjustments. For example, when agricultural user water consumption exceeds predicted values ​​for several consecutive days, the system can automatically increase the weight of agricultural demand, optimizing the allocation ratio for the next few days. This online learning capability enables the dynamic model to continuously adapt to structural changes in the system brought about by urban development, climate change, and technological advancements. This application achieves refined modeling and response to multiple dynamic characteristics of reclaimed water systems. By employing a hierarchical dynamic model framework, it can independently characterize the three core processes of supply and demand balance, water quality evolution, and user needs, and achieve overall coordinated optimization through data fusion and feedback mechanisms. Therefore, it effectively solves the problems of inaccurate predictions and rigid control caused by the simple structure of traditional dynamic models, improving the intelligence and robustness of the reclaimed water scheduling system. In particular, the output scheme covers multiple dimensions such as allocation quantity, timing, water quality assurance, and emergency response, enhancing the completeness and practicality of scheduling decisions, and is suitable for long-term planning and real-time operation management scenarios of complex reclaimed water networks in large and medium-sized cities.

[0045] Based on the above preferred embodiment, this embodiment further proposes that the three-layer dynamic structure also includes: a supply and demand dynamic layer representing a dynamic feedback dynamic model of reclaimed water supply and demand constructed based on dynamic theory, simulating the changes in supply and demand balance at different time scales; a water quality evolution layer representing a hydrodynamic-water quality coupled dynamic model simulating the spatiotemporal evolution of water quality indicators during the transmission and distribution of reclaimed water; and a user demand fluctuation layer representing a user demand prediction dynamic model established based on historical water use data and using time series analysis methods to capture short-term fluctuations and long-term trends in demand. Specifically, the supply and demand dynamic layer simulates the changing trends of the reclaimed water supply and demand balance by setting key variables such as population growth rate, industrial structure adjustment coefficient, and the degree of promotion of water-saving technologies; the water quality evolution layer uses water quality parameters from the comprehensive dataset as initial conditions, combined with regional hydrogeological characteristics, pollutant degradation coefficients, and water body self-purification capacity, to simulate the evolution patterns of water quality indicators such as COD (chemical oxygen demand), ammonia nitrogen (NH3-N), and total phosphorus (TP); and the user demand fluctuation layer analyzes historical water use data from industrial, agricultural, and municipal miscellaneous users to identify demand fluctuation cycles and their influencing factors, and establishes a user demand prediction dynamic model. The supply-demand dynamic layer refers to a nonlinear feedback mechanism between the supply and demand of reclaimed water, constructed based on system dynamics or differential equation modeling methods. This layer introduces quantifiable and controllable key socio-economic drivers—such as population growth rate (reflecting changes in domestic water consumption), industrial structure adjustment coefficient (reflecting the impact of water-intensive industry transformation on reclaimed water demand), and the degree of water-saving technology promotion (measuring the improvement in water efficiency per unit of output)—to form a set of state and control variables, used to characterize the evolution path of the supply-demand relationship of reclaimed water across multiple time scales, including daily, monthly, and annually. For example, in rapidly urbanizing areas, the effect of policy intervention can be simulated by increasing the parameter value of the degree of water-saving technology promotion, and its driving effect on the growth of reclaimed water reuse can be predicted. As an optional implementation, the precipitation change rate under climate change scenarios can also be introduced as an exogenous perturbation term to enhance the dynamic model's response to future extreme climate events. The water quality evolution layer employs a hydrodynamic-water quality coupled dynamic model, combining water flow equations (such as simplified forms of the shallow water equation or the Navier-Stokes equation) with water quality migration and transformation equations to jointly simulate the physical transport and biochemical reactions of reclaimed water during its flow in pipe networks or natural waterways. This layer uses effluent water quality parameters (such as initial concentrations of COD, ammonia nitrogen, and total phosphorus) obtained from a comprehensive dataset as boundary input conditions, combined with regional hydrogeological characteristics (including pipe flow velocity, water residence time, and topographic slope) and pollutant degradation coefficients (first-order reaction rate constants k1~k). nThe system calculates the water quality degradation and transformation patterns at each node and time period, using environmental parameters such as the water body's self-purification capacity (e.g., dissolved oxygen recovery rate, microbial degradation activity). In an optional embodiment, when applied to groundwater recharge-type reclaimed water utilization scenarios, a soil adsorption module and a groundwater seepage dynamics model can be added to expand the assessment function for nitrate nitrogen penetration risk.

[0046] The user demand fluctuation layer models long-term historical water use data from three typical user groups—industrial, agricultural, and municipal miscellaneous users—using time series analysis methods (such as ARIMA, SARIMA, or LSTM neural networks) to extract seasonal, periodic, and trend components. Through cluster analysis of water use behavior patterns among different user groups, key influencing factors are identified, such as industrial production cycles, crop irrigation seasons, and urban greening spraying frequencies, and these are encoded as input variables for the dynamic model, thus achieving dual capture of short-term sudden water peaks and long-term structural demand changes. An alternative approach is to introduce external data sources (such as weather forecasts, holiday schedules, and industrial operating rate statistics) to further improve prediction accuracy through multivariate regression or machine learning fusion strategies. These three sub-layers focus on the three core dimensions of quantity, quality, and demand, respectively, and are logically progressive and interactive with each other: the supply and demand dynamic layer provides a macro-level resource allocation framework, the water quality evolution layer ensures the bottom line of water supply security, and the user demand fluctuation layer supports refined scheduling decisions. Together, they constitute a multi-dimensional, collaborative dynamic simulation system, enabling the entire optimization allocation process to not only rely on static rule matching but also possess dynamic deduction and adaptive adjustment capabilities. This application achieves in-depth modeling of the complex evolutionary mechanisms within reclaimed water systems. By introducing mathematical expressions with clear physical meaning and statistical basis, it solves the technical problems of unclear evolutionary mechanisms of subsystems and low reliability of simulation results in traditional scheduling dynamics models, thus improving the applicability and robustness of the dynamics model under different regional conditions. In particular, by setting adjustable parameters (such as degradation coefficient, self-purification capacity, and industrial structure adjustment coefficient), the dynamics model can flexibly adapt to the reclaimed water management needs under different geographical, economic, and ecological backgrounds, thereby providing strong theoretical support and technical tools for formulating scientific, precise, and sustainable regional reclaimed water utilization strategies.

[0047] Based on the preferred embodiment described above, this embodiment further proposes: initializing the dynamic model and measuring the difference between the dynamic model output and actual demand to guide the dynamic model parameters; optimizing the difference and iterating the dynamic model parameters through multiple rounds to gradually approach the optimal solution; simulating and training the dynamic model based on the updated dynamic model parameters to obtain training results; and conducting verification tests based on the training results to obtain the tested dynamic model. Specifically, initializing the dynamic model refers to setting its initial structure and parameter configuration before deployment. This process includes determining the data dimensions of the input layer, the topological structure of each sub-dynamic model in the dynamic layer (such as the number of state variables in the supply and demand dynamic layer, the spatial discretization granularity of the water quality evolution layer, and the time window length of the user demand fluctuation layer), and the response format of the output layer. During initialization, the dynamic model parameters can be preset based on historical statistical data or expert experience. For example, the initial value of the population growth rate can be set to the average growth rate of the past five years, the industrial structure adjustment coefficient can be referenced from regional economic yearbook data, and the degree of water-saving technology promotion can be assigned based on policy coverage. Furthermore, the pollutant degradation coefficient in the water quality evolution layer can be initially estimated using the measured attenuation rate of a typical urban reclaimed water reuse system, while the water body's self-purification capacity is quantified by combining regional hydrogeological conditions (such as flow velocity, dissolved oxygen levels, and microbial activity). Measuring the difference between the dynamic model output and actual demand is a crucial step in evaluating the prediction accuracy of the dynamic model. This difference can be quantified by constructing a loss function; commonly used indicators include mean squared error (MSE), mean absolute percentage error (MAPE), or weighted comprehensive deviation index. Actual demand data comes from the historical operation records of the regional water resources dispatch center, covering information such as the actual allocation volume, water quality compliance rate, and emergency dispatch frequency for each water supply area during different time periods. By comparing the optimized configuration scheme output by the dynamic model with the actual operation data, prediction biases in specific scenarios (such as peak water use during the dry season or periods of concentrated industrial emissions) can be identified, thereby pinpointing unreasonable dynamic model parameters. For example, if the dynamic model consistently overestimates the reclaimed water utilization rate in agricultural irrigation areas, it may indicate insufficient learning of seasonal water use patterns in the user demand fluctuation layer, requiring adjustment of the periodic term weights in the time series dynamic model.

[0048] By optimizing discrepancies and iterating the kinetic model parameters multiple times, the aim is to achieve adaptive parameter correction. This process typically employs intelligent optimization strategies such as gradient descent, genetic algorithms, or Bayesian optimization to minimize prediction errors while ensuring the stability of the kinetic model. In each iteration, the system adjusts key variables based on the current loss value, such as correcting the influence weight of the industrial structure adjustment coefficient in the supply and demand dynamic layer, or recalibrating the kinetic constants of COD (chemical oxygen demand) degradation reactions in the water quality evolution layer. As the number of iterations increases, the kinetic model gradually converges to a set of parameters that minimizes the overall error. In particular, introducing a momentum term or an adaptive learning rate mechanism can accelerate the convergence speed and avoid getting trapped in local optima. The kinetic model is then trained using simulations based on the updated kinetic model parameters to accumulate operational experience under complex conditions. The training phase uses historical datasets to drive the kinetic model's operation, covering various typical scenarios, such as extreme weather events, temporary water restrictions during major municipal engineering projects, and emergency dispatching under sudden pollution accidents. In this process, the kinetic model not only learns the resource allocation patterns under normal conditions but also enhances its response capability to abnormal disturbances. The training results show that the internal parameters of the dynamic model tend to stabilize, and a set of decision-making logic with generalization ability is formed. For example, after experiencing multiple scenarios of a surge in ecological landscape water use due to high summer temperatures, the user demand fluctuation layer can more accurately capture such short-term peak characteristics. Validation testing is conducted based on the training results to verify the performance of the dynamic model on unseen data. The validation test set should be independent of the training data and include actual operational data for future time periods or different geographical regions. Test content includes, but is not limited to: the accuracy of reclaimed water allocation prediction, the rationality of water supply timing arrangements, the effectiveness of water quality protection measures, and the feasibility of emergency dispatch plans. If the validation results show that key indicators meet predetermined thresholds (e.g., MAPE ≤ 10%), the dynamic model is considered to have passed the test and become a tested dynamic model that can be used for actual dispatch. Otherwise, it returns to the parameter optimization stage for further iterative improvement. There is a close logical progression between each step: initialization provides the starting point for the dynamic model, error measurement reveals the problems, parameter optimization drives the evolution of the dynamic model, simulation training improves practical capabilities, and finally, validation testing confirms its reliability. The entire process forms a closed-loop feedback system, transforming the dynamic model from a statically deployed mathematical tool into an intelligent system with continuous learning and self-improvement capabilities. This application realizes the transformation of the dynamic model from theoretical conception to engineering usability. Because a complete training closed loop—initialization, error assessment, parameter optimization, training, and verification—is established, the problem of inaccurate predictions caused by large initial setting deviations and fixed parameters in traditional dynamic models is solved, thus significantly improving the accuracy and robustness of the reclaimed water optimization configuration scheme.Especially when facing uncertainties such as fluctuations in regional economic development, the impact of climate change, and changes in user behavior, this training mechanism gives the dynamic model a stronger adaptability, ensuring its long-term effective service to the regional reclaimed water dispatch system.

[0049] Based on the above preferred embodiment, this embodiment further proposes: analyzing the user type, water demand, and water quality requirements in the optimized configuration scheme to obtain the analysis results; dividing the reclaimed water supply area according to the analysis results; wherein, based on the analysis results and combined with water quality index data, determining the water supply priority, and planning four water supply areas with different water supply levels as the center: Level 1 water supply area, Level 2 water supply area, Level 3 water supply area, and Level 4 water supply area; establishing a mapping relationship table between reclaimed water use and water supply strategy, wherein industrial users correspond to a priority water supply strategy, and agricultural irrigation uses... The system employs a time-sharing water supply strategy for individual users and a dynamic water level maintenance strategy for users in the ecological landscape. It also sets corresponding triggering strategies for water supply based on four different water supply areas. Real-time monitoring of the effluent quality from the reclaimed water plant and the pressure at the end of the pipe network is conducted. When water quality indicators deviate from the user's required threshold, the deep treatment unit is automatically activated. When the pipe network pressure is lower than the set value, the frequency conversion regulation system of the booster pump station is triggered. Simultaneously, users within the area are guided according to the water supply strategy. Based on user feedback and operational energy consumption data, the water supply strategy is periodically optimized, the parameter weights in the mapping table are updated, and external resources are coordinated to supplement areas with insufficient water supply. The user's required threshold is set based on national or industry standards (such as "Water Quality Standard for Urban Wastewater Reclaimed Water for Landscape Environment (GB / T18921)") and in conjunction with user contract agreements. For example, the COD threshold for reclaimed water used for landscape replenishment is set to ≤50mg / L.

[0050] This paper analyzes and optimizes the user types, water demand, and water quality requirements in the reclaimed water allocation scheme to obtain the analysis results. The process of analyzing the optimized allocation scheme refers to extracting key scheduling instructions from the reclaimed water allocation scheme output by the dynamic model. This process reads the allocation scheme file through a data interface and uses natural language processing or structured parsing techniques to identify the user categories (e.g., industrial, agricultural, municipal miscellaneous uses), the water demand (unit: m³ / d) for each area, and the corresponding water quality standards (e.g., COD≤50mg / L, ammonia nitrogen≤10mg / L). This information forms the basis for subsequent zoning and strategy matching. User types are used to determine water use behavior characteristics and response priorities, water demand reflects the load scale, and water quality requirements determine the available water source level. For example, in an application scenario in a city in the middle and lower reaches of the Fen River, industrial park users are marked as having high water quality demand and continuous water use patterns, while farmland irrigation areas exhibit seasonal peaks and low water quality tolerance. Specifically, the city faces water scarcity and has great potential for reclaimed water utilization, but traditional scheduling methods are difficult to cope with dynamic changes. This dynamic model is used for reclaimed water optimization scheduling. A water source dataset is constructed by collecting and integrating remote sensing data on water surface area, flow data measured by on-site sensors, and water quality data from mobile rapid detection devices. A comprehensive dataset is then formed by combining real-time economic activity data (such as industrial electricity load and traffic flow) and ecological environment status data (such as temperature, precipitation, and vegetation index). This comprehensive dataset is input into a dynamic model to predict optimal reclaimed water allocation schemes. During the high-temperature summer months, the model predicts increased demand for landscape water and recommends pre-stocking high-quality reclaimed water for replenishing park lakes, while also initiating dynamic water level maintenance strategies. During the dry winter months, the model recommends prioritizing continuous industrial production water use, reducing unnecessary municipal water use, and triggering time-of-use water supply strategies. In the event of sudden heavy rainstorms, the model promptly adjusts the allocation scheme and activates emergency dispatch plans (such as activating backup water sources and enhancing deep treatment) to prevent water quality exceeding standards and water supply interruptions. This results in an increase in reclaimed water utilization rate from 20% to 35%, improved supply-demand matching accuracy, improved water quality compliance rate, and stable pressure at the end of the pipe network above the safe threshold (0.20 MPa). User satisfaction is enhanced, and operating energy consumption is reduced by more than 10%. The system's adaptive adjustment capability is strengthened, enabling real-time response to changes in regional demand. One optional implementation is to use a rule-based parsing module with preset configuration template formats to automatically match and extract fields. Another variation involves introducing a machine learning classifier to train a dynamic model to identify scheduling elements in unstructured text, improving adaptability to heterogeneous inputs.

[0051] Based on the analysis results, the reclaimed water supply area is divided. This division is based on geographical distribution, pipeline topology, and user clustering characteristics, dividing the entire service area into several functionally independent sub-regions. The division process comprehensively considers factors such as topographic elevation, pipeline routes, and the location of major water-using nodes to ensure that each region has a relatively independent water supply path and control capabilities. For example, administrative division layers and pipeline data can be loaded into a GIS system, and clustering algorithms (such as K-means) can be used to group adjacent users with similar water usage characteristics into the same region. This step also supports a dynamic adjustment mechanism—when a large-scale water-using project is added or a major pipeline renovation occurs, the system can recalculate the optimal partition boundaries to maintain scheduling flexibility. As an alternative, a fixed grid method can be used for region segmentation, dividing the entire area into equal-sized rectangular grids according to latitude and longitude, and then merging adjacent cells according to their dominant use. This method is suitable for rapid deployment in the initial modeling stage. Based on the analysis results and water quality data, water supply priorities were determined, and four water supply zones were planned, centered on the water supply area, with different water supply levels: Level 1, Level 2, Level 3, and Level 4. Water supply priority is a crucial parameter for assessing the order in which each area receives water supply security under water scarcity conditions. Its determination relies on multi-dimensional evaluation factors, including but not limited to: the economic contribution of user type (e.g., GDP share), the safety sensitivity of water use processes (e.g., whether production interruption risks are involved), ecological and environmental value (e.g., the impact of wetland replenishment on biodiversity), and historical compliance rates. Level 1 water supply zones typically refer to key national industrial projects or drinking water source replenishment areas, enjoying the highest priority for dispatch. Level 4 water supply zones may be for non-essential uses such as landscaping or road watering, and supply can be suspended during water shortages. The establishment of water supply level zones not only reflects administrative management intentions but also incorporates the resource matching principle of high-use areas for high-demand water and low-use areas for low-demand water. For example, reclaimed water that has undergone advanced treatment to meet the Class IV surface water standard is preferentially allocated to industrial cooling systems in the first-level zone, while effluent that has only undergone secondary treatment is used for urban greening irrigation in the third-level zone. This hierarchical system can be presented through a visualization platform to help decision-makers intuitively grasp the resource allocation pattern. An alternative implementation is to introduce fuzzy comprehensive evaluation to construct a priority scoring dynamic model, assigning adjustable weights to different factors to enhance the flexibility of policy guidance; another variation is to simulate the competitive relationships among multiple stakeholders based on a game theory framework to dynamically generate a fair and reasonable priority sequence.

[0052] A mapping table is established between reclaimed water uses and water supply strategies. Industrial users are assigned a priority water supply strategy, agricultural irrigation users a time-sharing water supply strategy, and ecological landscape users a dynamic water level maintenance strategy. Corresponding triggering strategies are set for four different water supply areas. This mapping table is a structured strategy configuration database that records the control logic and execution actions associated with different user types or water supply levels. The table is stored in a two-dimensional format, with rows representing user / area categories and columns including fields such as target use, recommended strategy, triggering conditions, execution equipment, and response time limit. For example, industrial users, due to their continuous production, are configured with a priority water supply strategy, ensuring their basic water consumption is not reduced during the scheduling cycle; agricultural users, significantly affected by crop growth cycles, are suitable for a time-sharing water supply strategy, reducing water supply intensity at night or during non-peak irrigation periods to stagger peak flows; ecological landscape users, such as artificial lakes and river replenishment points, adopt a dynamic water level maintenance strategy, adjusting the inflow in real time through level sensors to maintain visual appeal and ecological stability. Furthermore, this mapping relationship can be supplemented with differentiated parameters based on regional levels: industrial users in Level 1 zones maintain full water supply even when the overall network pressure is low, while similar users in Level 4 zones may receive intermittent water supply. This mechanism supports expansion to more strategy types, such as emergency flow restriction strategies for sudden pollution events and energy-saving strategies for reducing non-critical loads during peak electricity periods. In addition, the mapping table can be designed as an editable configuration file, allowing administrators to manually update the strategy binding relationships according to actual conditions. The system monitors the effluent quality of the reclaimed water plant and the pressure at the end of the pipeline network in real time. When water quality indicators deviate from the user's required threshold, the deep treatment unit is automatically activated. When the pipeline pressure is lower than the set value, the variable frequency control system of the booster pump station is triggered. Simultaneously, users within the area are guided according to the water supply strategy. The set value of the pipeline pressure is determined based on the pipeline hydraulic dynamics model and the safe pressure to ensure the water demand of users at the most unfavorable point; for example, it is set to 0.20 MPa. Real-time monitoring relies on a network of online monitoring equipment deployed at the reclaimed water plant outlet, key nodes of the main pipeline, and end users to collect key water quality parameters such as pH, turbidity, conductivity, residual chlorine, COD, and ammonia nitrogen, as well as operational status data such as pipeline pressure, flow rate, and temperature. The data is uploaded to the central control platform via the SCADA system, with a sampling frequency of up to once every 5 minutes to ensure timely detection of abnormal fluctuations. The automatic activation of the deep treatment unit means that when a water quality indicator is detected to exceed the target user's allowable range (e.g., COD > 60 mg / L while the user requires ≤ 50 mg / L), the control system immediately activates three-stage treatment modules such as ozone oxidation, activated carbon adsorption, or membrane filtration to enhance the purification effect. This action requires no manual intervention, with a response time of less than 30 seconds, effectively preventing substandard water from entering the user's end. The triggering of the booster pump station's variable frequency control system automatically adjusts the pump motor frequency when the pressure at the end of the pipeline network falls below a preset safety lower limit (e.g., 0.2 MPa), gradually increasing the head until stable water pressure supply is restored.Variable frequency drive (VFD) control significantly reduces energy consumption and mechanical shock compared to traditional start-stop control, extending equipment lifespan. User guidance is provided to affected users through smart terminal push notifications, factory LED screen displays, or remote valve opening adjustments, offering water-saving suggestions or temporarily adjusting water usage plans. For example, agricultural users may be notified of insufficient water pressure and advised to reduce irrigation area, or industrial enterprises may be advised to switch to an alternate water source.

[0053] One improvement is to integrate edge computing technology, enabling some judgment and response to be completed locally by the field controller, thus improving system robustness. Another optional implementation is to introduce a digital twin dynamics model to simulate the best response path under different fault scenarios, improving the scientific nature of decision-making. Based on user feedback and operational energy consumption data, the water supply strategy is regularly optimized, the parameter weights in the mapping table are updated, and external resources are coordinated to supplement areas with insufficient water supply. User feedback comes from multiple channels, including user satisfaction surveys, customer service work order systems, mobile APP problem reporting, and automated complaint identification systems, covering aspects such as water supply stability, sensory experience of water quality, and service response speed. Operational energy consumption data records operating cost indicators such as power consumption, chemical dosage, and equipment start-up and shutdown frequency for each pumping station, treatment unit, and chemical dosing system. Through statistical analysis and trend modeling of these two types of data, the system can identify the shortcomings of the existing water supply strategy. For example, if an agricultural area frequently reports nighttime water outages but energy consumption data shows that the pumping station is operating at full capacity for extended periods, it indicates that the time window setting of the time-sharing water supply strategy is unreasonable and the water supply period needs to be adjusted. For example, if a region is classified as a secondary water supply area but has a much higher user complaint rate than other similar regions, its priority weight should be reassessed. Updating the parameter weights in the mapping table means dynamically adjusting the relative importance of each influencing factor in the strategy selection dynamic model, such as increasing the coefficient of user satisfaction in priority calculation, or increasing the threshold sensitivity of energy consumption per unit of water in the energy-saving strategy triggering conditions. Coordinating external resources refers to initiating a cross-regional allocation mechanism when local reclaimed water supply cannot meet demand, introducing surplus capacity from nearby water plants, or temporarily activating emergency groundwater wells, mobile water purification devices, etc., to supplement water sources. This process can achieve resource sharing and joint scheduling through a regional water affairs collaboration platform. An alternative implementation is to use reinforcement learning algorithms to train the strategy optimization dynamic model, with minimizing the comprehensive cost (including energy consumption, complaints, and breach of contract compensation) as the objective function, and autonomously exploring the optimal strategy combination; another variation is to establish an expert knowledge base, combining human experience and data analysis results to jointly guide strategy iteration. This application realizes a closed-loop control process from macro-optimization configuration to micro-precise execution. By developing differentiated water supply strategies for different user types and regional levels, and establishing a dynamic control mechanism of real-time monitoring, automatic response, and continuous optimization, the system has solved problems such as response lag, rigid strategies, and resource mismatch in traditional scheduling. This achieves the technical effects of improving reclaimed water utilization efficiency, ensuring water security for key users, and reducing system energy consumption. In particular, the intelligent matching of usage and strategy through a mapping table makes scheduling instructions more adaptable to different scenarios; while the feedback-based strategy self-learning capability ensures the system operates at high efficiency for a long time, truly achieving the refined management goals of controllable, accurate, and optimized operation.

[0054] like Figure 2 As shown, a dynamic-based regional reclaimed water utilization scheduling system includes: Data acquisition module: It is used to collect data on natural water quantity, water quality, and water supply and demand to obtain water source datasets; Potential assessment module: It is used to analyze the water source dataset to assess the current status of reclaimed water utilization and determine the potential for reclaimed water utilization in conjunction with the target file; Optimization configuration module: It is used to integrate the reclaimed water utilization potential and regional dynamic characteristic data into a comprehensive dataset, and input the comprehensive dataset into the dynamic model to predict the optimal configuration scheme of reclaimed water; Water supply strategy module: It is used to trigger the corresponding water supply strategy according to the optimized configuration scheme and the purpose of reclaimed water.

[0055] The output of the data acquisition module is connected to the potential assessment module; the output of the potential assessment module and the regional dynamic characteristic data are input to the optimization configuration module; the optimized configuration scheme of reclaimed water output by the optimization configuration module is finally received and executed by the water supply strategy module.

[0056] The system of this invention includes a memory, a processor, and a dynamics-based regional reclaimed water utilization scheduling method program stored in the memory and executable on the processor. When executed by the processor, this program implements all the steps of the aforementioned method. The system's hardware architecture includes a data acquisition module, a potential assessment module, an optimization configuration module, and a water supply strategy module. These modules are connected via a high-speed communication bus to ensure real-time and reliable data transmission. For example, the data acquisition module acquires real-time monitoring data through a wireless sensor network and transmits it to the potential assessment module for analysis and processing; the potential assessment module sends the analysis results to the optimization configuration module, which generates an optimized configuration scheme using a dynamic model; the water supply strategy module then formulates a specific water supply strategy based on the optimized configuration scheme and disseminates relevant information to users within the region through user terminals.

[0057] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. A dynamic-based regional reclaimed water utilization scheduling method, characterized in that, include: Collect data on natural water quantity, water quality, and water supply and demand to obtain a water source dataset; The data set of water sources was analyzed to assess the current status of reclaimed water utilization and the potential for reclaimed water utilization was determined in conjunction with the target document; The reclaimed water utilization potential and regional dynamic characteristic data are integrated into a comprehensive dataset, and the comprehensive dataset is input into the dynamic model to predict the optimal configuration scheme of reclaimed water. The dynamic model includes: an input layer that receives the pre-processed integrated set; a dynamic layer with three layers to simulate changes in reclaimed water supply and demand, water quality evolution, and fluctuations in user demand; the three dynamic layers include a supply and demand dynamic layer, a water quality evolution layer, and a user demand fluctuation layer; and an output layer that outputs an optimized reclaimed water configuration scheme based on the simulation results of the dynamic layer. The corresponding water supply strategy is triggered based on the optimized configuration scheme and the intended use of reclaimed water.

2. The dynamic-based regional reclaimed water utilization scheduling method according to claim 1, characterized in that, Data on natural water quantity, water quality, and water supply and demand were collected to obtain a water source dataset, including: Adjust the parameters of the monitoring equipment according to environmental conditions and monitoring objectives; Real-time monitoring data is obtained based on the adjusted monitoring equipment; The real-time monitoring data is filtered to remove outliers and calibrated; at the same time, various monitoring data are synchronized to ensure that all types of data are aligned uniformly on the timeline. A water source dataset is obtained based on the real-time monitoring data.

3. The dynamic-based regional reclaimed water utilization scheduling method according to claim 1, characterized in that, The analysis of the aforementioned water source dataset assesses the current status of reclaimed water utilization and, in conjunction with the target document, determines the potential for reclaimed water utilization, including: Based on the aforementioned water source dataset, information on reclaimed water production capacity and demand distribution is extracted through multi-dimensional analysis. Construct an evaluation index system that combines local policies and objectives for reclaimed water utilization; Based on the aforementioned evaluation index system, a comprehensive score is given to the current status of reclaimed water utilization. Combined with regional economic development needs and ecological environment goals, the potential for reclaimed water utilization is determined.

4. The dynamic-based regional reclaimed water utilization scheduling method according to claim 3, characterized in that, Based on the aforementioned water source dataset, information on reclaimed water production capacity and demand distribution is extracted through multi-dimensional analysis, including: Based on the aforementioned water source dataset, key information is extracted from the dataset to obtain core data; Based on the core data, trend analysis is performed and historical data is used to predict future reclaimed water production capacity and demand distribution, resulting in predictions. Based on the prediction results and combined with the predefined reclaimed water utilization potential standards in the database, data mining and statistical analysis methods are used to assess the current treatment capacity, water quality compliance, utilization rate and distribution characteristics of the regional reclaimed water treatment facilities. Collect and interpret local standards, planning objectives and technical standards for reclaimed water utilization, in conjunction with regional economic development plans, population growth forecasts and environmental protection requirements; Further analysis of the alignment between the reclaimed water utilization guidelines and actual needs, and by comparing the gap between the current situation and the goals, identifies the constraints and development opportunities for reclaimed water utilization, ultimately determining the specific utilization potential of reclaimed water.

5. The dynamic-based regional reclaimed water utilization scheduling method according to claim 1, characterized in that, The reclaimed water utilization potential and regional dynamic characteristic data are integrated into a comprehensive dataset, including: Based on the aforementioned regional characteristic data, real-time economic activity and ecological environment status data for the day are obtained; The data on the potential for reclaimed water utilization and the real-time economic activities and ecological environment status are classified and coded to obtain coded data. The encoded data is then integrated to obtain a comprehensive dataset; The integrated dataset is preprocessed to obtain a preprocessed integrated set.

6. The dynamic-based regional reclaimed water utilization scheduling method according to claim 5, characterized in that, The dynamic model includes: Input layer: Receives the preprocessed aggregate set; Dynamic layer: A three-layer dynamic structure is set up to simulate changes in the supply and demand of reclaimed water, water quality evolution, and fluctuations in user demand; The three-layer dynamic structure includes a supply and demand dynamic layer, a water quality evolution layer, and a user demand fluctuation layer. Output layer: Based on the simulation results of the dynamic layer, output the optimized configuration scheme for reclaimed water; The optimized allocation scheme for reclaimed water includes the reclaimed water allocation volume, water supply sequence, water quality assurance measures, and emergency dispatch plan for each water supply area. By comparing actual operating data with predicted results through a feedback mechanism, the parameters of the dynamic model and the configuration strategy are dynamically adjusted to adapt to real-time changing needs.

7. The dynamic-based regional reclaimed water utilization scheduling method according to claim 6, characterized in that, The three-layer dynamic structure also includes: The supply and demand dynamic layer represents a dynamic feedback dynamic model of reclaimed water supply and demand constructed based on dynamic theory, simulating the changes in supply and demand balance at different time scales; The water quality evolution layer represents the use of a hydrodynamic-water quality coupled dynamic model to simulate the spatiotemporal evolution of water quality indicators during the transportation and distribution of reclaimed water; The user demand fluctuation layer is represented by a dynamic model for predicting user demand based on historical water usage data and using time series analysis to capture short-term fluctuations and long-term trends in demand. The supply and demand dynamic layer simulates the changing trend of the supply and demand balance of reclaimed water by setting key variables such as population growth rate, industrial structure adjustment coefficient, and the degree of promotion of water-saving technology. The water quality evolution layer uses the water quality parameters in the comprehensive dataset as initial conditions, and combines regional hydrogeological characteristics, pollutant degradation coefficients and water body self-purification capacity to simulate the evolution of water quality indicators such as chemical oxygen demand, ammonia nitrogen, and total phosphorus. The user demand fluctuation layer analyzes historical water usage data from industrial, agricultural, and municipal miscellaneous users to identify demand fluctuation cycles and their influencing factors, and establishes a user demand prediction dynamic model.

8. The dynamic-based regional reclaimed water utilization scheduling method according to claim 7, characterized in that, The dynamic model also includes dynamic model training and validation, as detailed below: The dynamic model is initialized and the difference between the dynamic model output and the actual requirements is measured to guide the dynamic model parameters; By optimizing the differences and iterating the dynamic model parameters multiple times, the dynamic adjustment gradually approaches the optimal solution; The dynamic model is simulated and trained according to the updated dynamic model parameters to obtain training results; The training results are used to perform validation tests to obtain the tested dynamic model.

9. The dynamic-based regional reclaimed water utilization scheduling method according to claim 1, characterized in that, Based on the optimized configuration scheme and the intended use of reclaimed water, the corresponding water supply strategy is triggered, including: The user type, water demand, and water quality requirements in the optimized configuration scheme are analyzed to obtain the analysis results. The reclaimed water supply area was divided based on the analysis results; Based on the analysis results and combined with water quality index data, water supply priorities were determined, and four water supply areas were planned with different water supply levels as the center: Level 1 water supply area, Level 2 water supply area, Level 3 water supply area, and Level 4 water supply area. Establish a mapping table between the uses of reclaimed water and water supply strategies, in which industrial users correspond to priority water supply strategies, agricultural irrigation users correspond to time-sharing water supply strategies, and ecological landscape users correspond to dynamic water level maintenance strategies. Also, set corresponding trigger water supply strategies according to the four different water supply areas. The system monitors the effluent quality of the reclaimed water plant and the pressure at the end of the pipeline network in real time. When the water quality indicators deviate from the user's required threshold, the deep treatment unit is automatically started. When the pipeline pressure is lower than the set value, the frequency conversion regulation system of the booster pump station is triggered. At the same time, the system guides users in the area according to the water supply strategy. Based on user feedback and operational energy consumption data, we regularly optimize water supply strategies, update parameter weights in the mapping table, and coordinate external resources to supplement areas with insufficient water supply.

10. A dynamic-based regional reclaimed water utilization scheduling system, used to execute the dynamic-based regional reclaimed water utilization scheduling method as described in any one of claims 1-9, characterized in that, include: Data acquisition module: It is used to collect data on natural water quantity, water quality, and water supply and demand to obtain water source datasets; Potential assessment module: It is used to analyze the water source dataset to assess the current status of reclaimed water utilization and determine the potential for reclaimed water utilization in conjunction with the target file; Optimization configuration module: It is used to integrate the reclaimed water utilization potential and regional dynamic characteristic data into a comprehensive dataset, and input the comprehensive dataset into the dynamic model to predict the optimal configuration scheme of reclaimed water; Water supply strategy module: It is used to trigger the corresponding water supply strategy according to the optimized configuration scheme and the purpose of reclaimed water.

Images