A multi-model coupling-based intelligent collaborative scheduling method and system for cascade pump stations and canal systems

Abstract

The application belongs to the technical field of water conservancy project operation scheduling, and discloses a cascade pump station-canal system intelligent collaborative scheduling method and system based on multi-model coupling, which is suitable for the Shandong main line project of the east line of the South-to-North Water Diversion. The method collects and pre-processes whole-process data, constructs a water balance analysis model and a one-dimensional water dynamic model, realizes deep coupling, solves an optimal scheduling scheme by using a multi-objective optimization algorithm based on project-specific constraint conditions, and forms a closed-loop control in combination with real-time feedback. The application innovatively realizes the integration and collaboration of monthly-scale water quantity prediction and hourly-scale hydraulic control, greatly improves the scheduling scientificity, safety guarantee capability and operation energy efficiency, and solves the technical problems of experience dependence, response lag and insufficient collaboration in traditional scheduling.

Description

Technical Field

[0001] This invention belongs to the field of water conservancy project operation and scheduling technology, and relates to water resource optimization and cross-regional water transfer project scheduling technology. Specifically, it is applicable to the joint optimization scheduling, intelligent control and digital twin application of the cascade pumping station group and open channel / pipeline water conveyance system of the Shandong trunk line of the South-to-North Water Transfer Project. Background Technology

[0002] The Shandong trunk line project of the first phase of the East Route of the South-to-North Water Diversion Project undertakes the important task of supplying water to multiple targets in the Shandong Peninsula, northern Shandong, Hebei and Tianjin. It faces multiple challenges, including complex and ever-changing water supply targets, long-distance water transmission, multi-stage layout, multiple constraints, and efficient operation of the cascade pumping station group.

[0003] Currently, the Shandong trunk line dispatching adopts a "segmented independent control + experience-based supplementation" model. The cascade pumping stations in the southern Shandong section focus on matching water lifting capacity, while the sluice gates in the northern Shandong and Jiaodong sections focus on water level control. The existing planned dispatching and model simulation are often disconnected, resulting in a high reliance on manual experience and trial and error in actual operation. This makes it difficult to respond quickly to sudden changes, and there is still considerable room for optimization in overall operational efficiency and safety. Summary of the Invention

[0004] This invention aims to solve the technical problems of poor connection and iterative lag between the long-term water balance plan driven by "water demand" and the short-term real-time scheduling and control centered on "operational safety" under the complex topology of the Shandong trunk line of the first phase of the East Route of the South-to-North Water Diversion Project. It achieves integrated and coordinated optimization from monthly water volume prediction to hourly hydraulic precision control.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: The intelligent collaborative scheduling method for cascade pumping stations and canal systems based on multi-model coupling of the present invention specifically includes the following steps: 1. Data Acquisition and Preprocessing The data collection covered the entire Shandong main canal section, including water volume monitoring data at the Jiangsu-Shandong provincial border, operational data from seven pumping stations in the southern Shandong section, gate and pumping station data in the northern Shandong / Jiaodong section, cross-sectional data from 52 sub-units across four major canal sections, hydrological and meteorological data, and water demand data from users at the water diversion gates. Preprocessing employed outlier removal, spatiotemporal interpolation, and data standardization methods to ensure data compatibility with the dual-model calculation requirements.

[0006] 2. Model Building and Deep Coupling - Water balance analysis model construction: Based on the monthly water demand of users at each water distribution point along the route and the progress of water transfer in the previous stage, the water supply, water replenishment and storage capacity of lakes and reservoirs are comprehensively considered. The flow capacity of pumping stations, control gates, water diversion gates and channels are used as constraints. The impact of special situations such as engineering maintenance, water transfer during the ice period and water pollution is also taken into account. According to the annual water volume scheduling plan, the water volume transfer volume at the Jiangsu-Shandong provincial boundary, the water volume transfer volume at key engineering nodes, and the monthly water volume allocation results of users at the water distribution points are calculated using the water balance equation. The monthly water transfer arrangements for the remaining period are also predicted.

[0007] - One-dimensional hydrodynamic model construction: Based on the distribution characteristics along the Shandong trunk line project, the physical properties of the channels, and the scheduling and management boundaries, the project is divided into four major sections: Hanzhuang section, Lianghu section, Jiaodong section, and Lubei section. These sections are further subdivided into 52 channel sub-units, with each sub-unit embedding the corresponding pumping station and sluice gate engineering parameters. The Saint-Venant equation is used to construct a hydraulic calculation model, which supports independent calculation or flexible assembly of sub-units and outputs the dynamic changes in water level and flow rate of each sub-unit.

[0008] - Deep coupling mechanism: Design coupling interface and iteration strategy. The coupling interface realizes bidirectional data transmission between the two models. The iteration strategy sets a convergence threshold and performs synchronous calculation iteration of the two models at a fixed step size until the result meets the convergence condition, forming a dynamic coupling calculation closed loop.

[0009] 3. Cooperative Scheduling Constraints and Optimization Solutions - Constraints: Four types of project-specific constraints are clearly defined, including pump station constraints (maximum / minimum head, flow range, and unit start-up / shutdown limits for the 7 pump stations in the Lunan section), canal system constraints (upper and lower limits of water levels and design flow for the 4 major dispatching units and 52 sub-units), water quantity constraints (the scope of the Jiangsu-Shandong provincial boundary water transfer agreement, the water supply guarantee rate of the water distribution gate, and the threshold for controlling water loss in the main line), and sluice gate constraints (the opening adjustment range, response speed, and design flow of sluice gates in the Lubei / Jiaodong section).

[0010] - Optimization Solution: A multi-objective optimization algorithm is adopted, with the optimization objectives of "maximum water supply guarantee rate, minimum water transmission energy consumption, and minimum water loss". Based on the calculation results of the coupled model, the optimal scheduling scheme is solved, including pump station operating parameters, gate opening degree, and water distribution adjustment value.

[0011] 4. Scheduling execution and real-time feedback adjustment The optimal scheduling scheme is sent to the control cabinets of each pump station and the controller of the gate station to realize the execution of the instructions; the engineering operation data is collected in real time through the data perception layer and compared with the prediction results of the coupled model. When the deviation exceeds the convergence threshold, the dual model is triggered to re-iterate the calculation and generate an adjustment scheme, realizing the closed-loop control of the whole process of "calculation-execution-monitoring-feedback-adjustment".

[0012] This invention also discloses an intelligent collaborative scheduling system for cascade pumping stations and canals based on multi-model coupling, including a data acquisition and preprocessing module, a model construction and coupling module, an optimization solution module, and a scheduling execution and feedback module. Each module corresponds to the functional implementation of the above-mentioned method steps, ensuring the effective implementation of the method.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are as follows: 1. Innovation in engineering adaptability: Specifically adapted to the Shandong section of the East Route of the South-to-North Water Diversion Project, the one-dimensional hydrodynamic model is refined into 52 channel sub-units, supporting independent calculation or flexible assembly, accurately matching the differences in physical attributes and engineering distribution of different channel sections, and breaking through the limitations of existing models that have strong versatility but weak engineering adaptability.

[0014] 2. Innovative Coordinated Scheduling Strategy: Construct a three-in-one coordinated strategy of "cascade pumping stations - segmented gate stations - water distribution", dynamically optimize pumping station water lifting parameters and gate station opening, match the fluctuation of provincial boundary water transfer volume with user water demand, and achieve linkage and adaptation of the entire trunk line operating conditions.

[0015] 3. Scenario-based Adaptation Innovation: Through sub-unit assembly operations and dynamic adjustment of coupling models, it supports flexible switching between different scheduling scenarios, quickly outputs scheduling schemes adapted to the scenarios, and improves the ability of engineering to cope with complex working conditions.

[0016] 4. Significantly improved scientific scheduling: The traditional "human experience + single model calculation" mode is upgraded to an intelligent decision-making mode of "dual model coupling drive", which enables scheduling instructions to simultaneously possess macro-level water quantity rationality and micro-level hydraulic safety.

[0017] 5. Safety and efficiency quantitative optimization: Through pre-simulation using hydrodynamic models, risks such as exceeding water level limits and excessive water flow velocity can be warned in advance, and the safety verification time for major scheduling operations can be shortened from several hours to minutes.

[0018] 6. Multi-objective optimization: Achieving comprehensive optimization of water supply guarantee rate, water transmission energy consumption, and water loss, significantly improving the energy efficiency of project operation and water resource utilization efficiency, and providing reliable technical support for intelligent scheduling of inter-regional water transfer projects. Detailed Implementation

[0019] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described below with reference to embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0020] Numerous specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways than those described herein, and therefore the invention is not limited to the specific embodiments disclosed in the following specification.

[0021] Example 1

[0022] The following detailed description of the implementation process of the intelligent collaborative scheduling method and system for cascade pumping stations and canals based on multi-model coupling, combined with the actual working conditions of the Shandong trunk line of the South-to-North Water Diversion Project (Eastern Route), ensures that those skilled in the art can reproduce the technical solution of the present invention.

[0023] I. Implementation Preparation (a) Confirmation of project background parameters This embodiment focuses on the entire process of the Shandong trunk line of the eastern route of the South-to-North Water Diversion Project, specifying the core engineering parameters: 1. Cascade pumping stations: 7 pumping stations in the southern Shandong section (in order: Taierzhuang Pumping Station, Wannianzha Pumping Station, Hanzhuang Pumping Station, Second-Level Dam Pumping Station, Changgou Pumping Station, Denglou Pumping Station, and Baliwan Pumping Station). The maximum head of a single pumping station is 38m and the minimum head is 12m. The design flow rate of a single unit is 50m³ / s. Each pumping station has 3-5 units. The maximum number of start-ups and shutdowns of a unit per day is limited to 2. 2. Canal System Division: The canal system is divided into four major dispatching units: Hanzhuang section (128km long), Lianghu section (including Nansihu and Dongpinghu, 95km long), Jiaodong section (289km long), and Lubei section (172km long). These units are further subdivided into 52 canal sub-units, each ranging from 3 to 15km in length. The sub-unit cross-section is trapezoidal or rectangular, with a slope coefficient of 1:2.5-1:3.0. The canal design water depth is 3-5m, with the upper limit of the water level set at a safety freeboard of 0.5m and the lower limit of the water level set at a minimum submersion depth of 1.2m below the pump station inlet. 3. Distribution of gate stations: A total of 23 control gates and 18 water diversion gates are set up in the Lubei section and Jiaodong section. The gate station opening adjustment range is 0-100%, the adjustment response speed is 0.5% / s, the maximum flow capacity of a single gate station is 80m³ / s, and the minimum flow capacity is 5m³ / s. 4. Water Constraints: The water transfer agreement between Jiangsu and Shandong provinces covers 10-15 million m³ / day. The water distribution gates cover 32 users, including industrial water users in the Shandong Peninsula, agricultural irrigation water users in northern Shandong, and domestic water users in cities along the route. The water supply guarantee rate is required to be ≥95%, and the water loss control threshold for the main line is ≤3% / 100km.

[0024] (II) Hardware and Software Environment Configuration 1. Hardware System: - Data acquisition equipment: Ultrasonic flow meters (measurement accuracy ±1%), radar water level gauges (measurement accuracy ±2cm), and meteorological stations (monitoring parameters such as air temperature, precipitation, and wind speed, with a sampling frequency of 1 time / hour) were deployed at the Jiangsu-Shandong provincial border, the inlet and outlet of each pumping station, the upstream and downstream of the sluice gate, key sections of the channel sub-units, and the water distribution gate. A total of 186 sets of monitoring equipment were deployed, and real-time data transmission was achieved through 4G / 5G IoT modules. - Computing equipment: It adopts an industrial server cluster (including 3 main servers and 6 compute node servers). The main servers are equipped with Intel Xeon Gold 6348 CPUs, 128GB of memory and 4TB of hard disk capacity. The compute node servers are equipped with Intel Xeon Silver 4314 CPUs and 64GB of memory, supporting parallel computing. - Control equipment: Each pump station control cabinet and gate controller adopts a PLC (Programmable Logic Controller, model S7-1500), which supports Modbus TCP communication protocol and can receive remote dispatching commands and provide feedback on equipment operating status.

[0025] 2. Software System: - Operating System: The server uses Linux CentOS 7.9, and the control terminal uses Windows 10 IoT. - Development environment: Python 3.9 programming language is used, combined with NumPy and Pandas libraries for data processing, TensorFlow framework is used to implement model coupling calculation, and Gurobi optimizer is used for multi-objective optimization solution; - Visualization Platform: Develop a B / S architecture scheduling and monitoring platform that supports real-time data display, visualization of model calculation process, and distribution and execution status feedback of scheduling schemes. Specific implementation steps

[0026] (I) Data Acquisition and Preprocessing Implementation 1. Data Acquisition: Using the aforementioned monitoring equipment, data is collected every 15 minutes, including the following: - Water transfer volume at the Jiangsu-Shandong provincial border (real-time flow data); - Unit operating status (operating / shutdown), actual head, output flow, and power consumption of the 7-stage pumping stations in the Lunan section; - Real-time opening degree, upstream and downstream water levels, and flow rate of the sluice gates and pumping stations in the Lubei / Jiaodong section; - Water level and flow velocity data for key sections of 52 channel sub-units; - Hydrometeorological data (daily precipitation, daily average temperature, daily average wind speed); - Real-time water intake flow rate of 32 water outlet gates and the user's remaining water demand for the day.

[0027] 2. Data preprocessing: - Outlier removal: The 3σ criterion is used to remove flow and water level data that exceed the range of "mean ± 3 times standard deviation". For example, if the water level monitoring value of a certain section is 6.8m, while the historical average value of the same period of the section is 4.2m and the standard deviation is 0.5m, 6.8m exceeds the range of 4.2 ± 3 × 0.5 = 2.7-5.7m, and is judged as an outlier and removed. - Spatiotemporal interpolation: Linear interpolation is used to supplement missing 15-minute data; Kriging interpolation is used to calculate water level and flow data of non-monitored sections of channel sub-units based on data from adjacent monitoring sections. - Data standardization: Convert data of different dimensions to the [0,1] interval. For example, the standardization formula for pump station head is: H'=(H-Hmin) / (Hmax-Hmin), where H is the actual head, Hmax=38m, Hmin=12m, to ensure that the data is adapted to the dual-model calculation requirements.

[0028] (II) Model Construction and Deep Coupling Implementation 1. Water balance analysis model construction and parameter setting: - Input parameters: monthly water demand of 32 water distribution gates (50,000-80,000 m³ / day for industrial users, 100,000-150,000 m³ / day for agricultural users, and 30,000-50,000 m³ / day for domestic users), cumulative water transfer volume in the previous period, initial water storage of Nansi Lake / Dongping Lake (set the initial water storage volume to 60% of the total reservoir capacity), engineering maintenance plan (e.g., monthly maintenance of Hanzhuang Pumping Station, maintenance duration of 3 days), and water transfer period during the ice season (late December to early February, water transfer capacity is calculated at 80% of the design flow). - Water balance equation: Q provincial boundary + Q lake and reservoir replenishment = Q water diversion along the route + Q water conveyance loss + ΔQ lake and reservoir storage, where Q provincial boundary is the water diversion volume at the Jiangsu-Shandong provincial boundary, Q lake and reservoir replenishment is the water replenishment volume of lakes and reservoirs, Q water diversion along the route is the total water intake volume of each water diversion gate, Q water conveyance loss = total length of the main line × water conveyance loss rate per unit length, and ΔQ lake and reservoir storage is the change in lake and reservoir storage (positive value is storage, negative value is release). - Calculation output: The monthly water transfer volume allocation at the Jiangsu-Shandong provincial border (e.g., 12 million m³ / day in October), the monthly average water transfer volume of each key engineering node (inlet and outlet of pumping station, upstream and downstream of sluice gate), and the monthly water use quota of each water distribution gate are calculated through the above equations. At the same time, the water transfer trend for the next 3 months is predicted.

[0029] 2. Construction and parameter setting of one-dimensional hydrodynamic model: - Sub-unit parameter embedding: Each channel sub-unit embeds the corresponding topographic data (cross-sectional dimensions, roughness coefficient n=0.018-0.022, adjusted according to the channel bed material) and pump station / sluice gate engineering parameters (such as embedding the head-flow characteristic curve of Hanzhuang pump station and the opening-flow capacity curve of Hanzhuang sluice gate in the Hanzhuang section sub-unit). - Solving the governing equations: The Saint-Venant equations are used (continuity equation ∂Q / ∂x + ∂A / ∂t = 0, momentum equation ∂Q / ∂t + ∂(Q² / A) / ∂x + gA∂h / ∂x + gQ|Q| / (C²AR) = 0), where Q is the flow rate, x is the distance along the channel, t is the time, A is the cross-sectional area of ​​the water passage, h is the water level, g is the gravitational acceleration, C is the Chezy coefficient, and R is the hydraulic radius; the governing equations are discretized using the Preissmann four-point implicit difference scheme, with a time step of 10 minutes and a spatial step of 500 m. - Calculation output: Outputs dynamic data of water level, flow rate, and velocity for each channel sub-unit every 10 minutes. For example, at 10:00, the water level of the third sub-unit of Hanzhuang section is 4.2m, the flow rate is 65m³ / s, and the velocity is 0.8m / s.

[0030] 3. Implementation of deep coupling between two models: - Coupling Interface Design: The coupling interface is constructed using the Socket communication protocol to realize bidirectional data transmission between the water balance analysis model and the one-dimensional hydrodynamic model. The transmission cycle is 1 hour. The transmitted data includes the monthly water allocation plan (split by hour) output by the water balance model and the real-time hydraulic parameters output by the hydrodynamic model. - Iterative strategy execution: Set convergence thresholds of water level deviation ≤ 0.1m and flow deviation ≤ 3m³ / s, and perform synchronous iterative calculations using dual models in 1-hour increments. For example, when the hydrodynamic model calculates a water level of 4.8m for a sub-unit, exceeding the upper limit of 4.7m for that sub-unit, a feedback signal is generated, and the hourly flow allocation of the water balance model is adjusted (from 60m³ / s to 55m³ / s). The flow is then re-input into the hydrodynamic model for calculation until the water level deviation is ≤ 0.1m, forming a coupled calculation closed loop.

[0031] (III) Loading and Optimization Solution of Cooperative Scheduling Constraints 1. Specific setting of constraints: - Pump station constraints: The operating head range of a single pump in the 7-level pump station in the Lunan section is 12-38m, the flow rate range of a single pump is 30-50m³ / s, the total output flow of the pump station does not exceed the total design flow of the unit (3-5 units × 50m³ / s), and the flow matching deviation between adjacent pump stations is ≤10m³ / s. - Canal system constraints: The upper limit of water level for 52 canal sub-units is 4.0-5.5m (adjusted according to the location of the sub-unit), the lower limit of water level is 2.0-3.2m, and the flow velocity range is 0.3-1.2m / s (to avoid siltation due to excessively low flow velocity and scouring of the canal due to excessively high flow velocity). - Water volume constraints: The water diversion volume at the Jiangsu-Shandong provincial border is 10-15 million m³ / day, and the water supply guarantee rate of each water distribution gate is ≥95% (i.e., the actual monthly water supply is not less than 95% of the monthly demand), and the water loss in the main line is ≤3% / 100km (for example, the water loss in the 100km canal does not exceed 3% of the input flow). - Gate constraints: The gate opening adjustment range of the Lubei section / Jiaodong section is 0-100%, the single adjustment amplitude is ≤10%, the adjustment response time is ≤20s, and the gate flow rate does not exceed the design flow rate of 80m³ / s.

[0032] 2. Multi-objective optimization solution: - Objective function construction: 1. Maximize water supply guarantee rate: max f1 = Σ(actual water supply / demanded water supply) / n, where n is the number of water distribution gates (32). 2. Minimize water conveyance energy consumption: min f2 = Σ (energy consumption of a single pump unit × number of operating units × operating time), the formula for energy consumption of a single pump unit is E = k × H × Q × t (k is the energy consumption coefficient, with a value of 0.75-0.85 kWh / (m³·m), H is the head, Q is the flow rate, and t is the operating time). 3. Minimize water loss: min f3 = Σ (water transfer loss / total water transfer volume); - Solution process: The multi-objective optimization is transformed into single-objective optimization using the weighted summation method. The weights of f1, f2, and f3 are set to 0.4, 0.3, and 0.3, respectively. The optimal scheduling scheme is obtained by solving the problem using the Gurobi optimizer. This scheme includes: the unit operation combination of the 7 pumping stations in the Lunan section (e.g., 2 units operating at Taierzhuang Pumping Station and 3 units operating at Wannianzha Pumping Station), the output flow of a single unit (35-50 m³ / s), the opening degree of each sluice gate in the Lubei / Jiaodong section (30%-85%), and the hourly water intake adjustment of each branch gate (within ±5%).

[0033] (iv) Implementation of scheduling execution and real-time feedback adjustment 1. Dispatch scheme issuance: The dispatch instructions (pump station unit start / stop instructions, flow setpoints, gate opening adjustment values) obtained from the optimization solution are issued to the PLC control cabinets and gate controllers of each pump station via industrial Ethernet. The issuance delay is ≤5s, and the instruction transmission adopts an encryption protocol to ensure security.

[0034] 2. Real-time Monitoring and Deviation Judgment: The data perception layer collects actual operational data of the project every 15 minutes (actual flow rate, head of pumping stations, actual opening of sluice gates, actual water level in channels, etc.), compares it with the prediction results of the coupled model, and calculates the deviation value. For example, if the model predicts that the output flow rate of Hanzhuang Pumping Station is 45 m³ / s, and the actual collected value is 42 m³ / s, the flow deviation is 3 m³ / s, which does not exceed the convergence threshold (3 m³ / s), and the current scheduling plan is maintained; if the actual collected water level is 4.9 m, which exceeds the model prediction value of 4.7 m, the deviation is 0.2 m, which exceeds the water level convergence threshold (0.1 m), triggering the feedback adjustment mechanism.

[0035] 3. Model Re-iteration and Scheme Adjustment: When the deviation exceeds the threshold, the actual operating data is used as the new input condition to trigger the water balance analysis model and the one-dimensional hydrodynamic model to re-iterate and calculate. The iteration step is shortened to 5 minutes, and an adjustment scheme is quickly generated (such as reducing the upstream pump station flow rate from 45m³ / s to 40m³ / s and adjusting the gate opening from 60% to 55%). The scheme is then issued for execution, realizing a closed-loop control of the entire process of "calculation-execution-monitoring-feedback-adjustment".

[0036] III. Implementation Results Verification This embodiment demonstrates a 30-day trial operation conducted on the Shandong trunk line of the South-to-North Water Diversion Project's eastern route, following the steps outlined above. The verification results are as follows: 1. Dispatch response speed: The safety verification time for major dispatch operations (such as the increase in water transfer volume at provincial borders from 12 million m³ / day to 14 million m³ / day) has been shortened from 4 hours in the traditional mode to 25 minutes, improving response efficiency by more than 90%; 2. Safety assurance capability: No safety risks such as water level exceeding the limit or abnormal flow velocity occurred during the trial operation. The water level control deviation was ≤0.08m and the flow control deviation was ≤2.5m³ / s, meeting the requirements for safe operation of the project. 3. Multi-objective optimization effect: The water supply guarantee rate reached 98.3%, which is 3.1 percentage points higher than the traditional model; the average energy consumption of water transmission was reduced by 12.7%; the water loss rate of trunk line transmission was controlled at 2.2% / 100km, which is 0.8 percentage points lower than the traditional model, realizing the comprehensive optimization of water resource utilization efficiency and engineering operation energy efficiency.

[0037] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments for application in other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for intelligent collaborative scheduling of cascade pumping stations and canal systems based on multi-model coupling, characterized in that, Includes the following steps: 1) Data Acquisition and Preprocessing: Collect full-process data of the Shandong main line of the South-to-North Water Diversion Project (Eastern Route), including water volume monitoring data at the Jiangsu-Shandong provincial border, operation data of 7 pumping stations in the southern Shandong section, gate station data in the northern Shandong / Jiaodong section, cross-sectional data of 52 sub-units in 4 major canal sections, hydrological and meteorological data, and water demand data of users at water distribution gates; use outlier removal, spatiotemporal interpolation, and data standardization methods to process the data; 2) Model Construction and Deep Coupling: A water balance analysis model is constructed. Based on monthly water demand, water transfer progress, engineering constraints, and the impact of special circumstances, the monthly water allocation results and subsequent water transfer arrangements are obtained through the water balance equation. A one-dimensional hydrodynamic model is constructed, divided into four major sections: Hanzhuang section, Lianghu section, Jiaodong section, and Lubei section, and further refined into 52 channel sub-units. Engineering parameters are embedded, and hydraulic simulation is achieved based on Saint-Venant's equations. Deep coupling between the two models is achieved through coupling interfaces and iterative strategies to form a dynamic calculation closed loop. 3) Coordinated scheduling constraints and optimization solutions: Define the specific constraints for four types of projects: pumping stations, canal systems, water volume, and sluice gates; adopt a multi-objective optimization algorithm to solve for the optimal scheduling scheme of pumping station operating parameters, sluice gate opening degree, and water volume allocation adjustment values ​​with the objectives of maximizing water supply guarantee rate, minimizing water transmission energy consumption, and minimizing water volume loss. 4) Scheduling execution and real-time feedback adjustment: Issue and execute the optimal scheduling plan, collect engineering operation data in real time and compare it with the model prediction results. When the deviation exceeds the convergence threshold, trigger the dual model to re-iterate and generate an adjustment plan to achieve closed-loop control.

2. The method according to claim 1, characterized in that, The constraints of the water balance analysis model described in step 2 include special cases such as the regulation capacity of lakes and reservoirs, the flow capacity of engineering projects, engineering maintenance, water transfer during ice periods, and water pollution.

3. The method according to claim 1, characterized in that, The one-dimensional hydrodynamic model described in step 2 supports independent calculation or flexible assembly of 52 channel sub-units, adapting to the differences in physical properties and engineering distribution of different channel sections.

4. The method according to claim 1, characterized in that, The iterative strategy described in step 2 sets a convergence threshold and performs synchronous calculations of the two models at a fixed step size until the result meets the convergence condition.

5. The method according to claim 1, characterized in that, The pump station constraints mentioned in step 3 include the maximum / minimum head, flow range, and unit start-up / shutdown frequency limits for the 7 pump stations in the Lunan section; the canal system constraints include the upper and lower limits of water levels and design flow for the 4 major dispatching units and 52 sub-units; the water volume constraints include the water volume agreement range of the Jiangsu-Shandong provincial boundary, the water supply guarantee rate of the water distribution gate, and the control threshold for water loss in the main line; and the gate station constraints include the gate station opening adjustment range, response speed, and design flow for the Lubei / Jiaodong section.

6. A multi-model coupling-based intelligent collaborative scheduling system for cascade pumping stations and canal systems, characterized in that, include: 1) Data Acquisition and Preprocessing Module: Used to collect full-process data of the Shandong trunk line, and output standardized data adapted to dual-model calculation through outlier removal, spatiotemporal interpolation, and data standardization. 2) Model building and coupling module: including water balance analysis model unit, one-dimensional hydrodynamic model unit and coupling unit. The coupling unit realizes bidirectional data transmission between the two models through the coupling interface and forms a dynamic coupling calculation closed loop through the iterative strategy. 3) Optimization and solution module: This module loads project-specific constraints and uses a multi-objective optimization algorithm to solve for the optimal scheduling scheme based on the calculation results of the coupled model. 4) Scheduling execution and feedback module: used to send the optimal scheduling plan to the field control equipment, collect operating data in real time and compare it with the model prediction results, and trigger deviation adjustment and model re-iteration.

7. The system according to claim 6, characterized in that, The data acquisition and preprocessing module covers a full range of data, including water volume at the Jiangsu-Shandong provincial border, pump station operation, sluice gate operation, channel cross-section, hydrological and meteorological data, and user water demand.

8. The system according to claim 6, characterized in that, The one-dimensional hydrodynamic model unit is divided into four major sections: Hanzhuang section, Lianghu section, Jiaodong section, and Lubei section, and further subdivided into 52 channel sub-units. Each sub-unit embeds the corresponding pumping station and sluice gate engineering parameters.

9. The system according to claim 6, characterized in that, The optimization objectives of the optimization solution module are to achieve the highest water supply guarantee rate, the lowest water transmission energy consumption, and the lowest water loss. The output results include pump station operating parameters, gate opening degree, and water distribution adjustment value.

10. The system according to claim 6, characterized in that, The scheduling execution and feedback module supports closed-loop control of the entire process of "calculation-execution-monitoring-feedback-adjustment", enabling dynamic optimization of the scheduling scheme.