A multi-gate coordinated control system and control method thereof

Abstract

A multi-gate collaborative control system and its control method relate to the field of water conservancy information automation and intelligent control. The system includes: real-time acquisition of multi-gate hydrological status data, and numerical validity testing and logical judgment; preprocessing of the logically judged multi-gate hydrological status data to obtain structured multi-gate hydrological status data; predicting the future hydrological status of each gate node based on the structured multi-gate hydrological status data to obtain structured hydrological status prediction results; generating three scoring indicators based on the structured hydrological status prediction results; weighting the three scoring indicators to obtain a comprehensive score; and generating a multi-gate collaborative optimal opening strategy based on the comprehensive score to control gate movement and collect multi-gate hydrological status data in real time to achieve a closed-loop system. This system addresses the problems of traditional gate control systems, such as a lack of a collaborative scheduling mechanism, insufficient prediction capabilities, delayed control response, and rigid regulation strategies.

Description

Technical Field

[0001] The present invention relates to the field of water conservancy information automation and intelligent control, and in particular to a multi-gate coordinated control system and a control method thereof. Background Art

[0002] With the continuous improvement of water conservancy informationization and automation, gate control, as a core link in river basin scheduling, water resources management, flood control and drainage, has become an important direction for the current system upgrade in terms of intelligent and coordinated control capabilities. Traditional gate control systems mostly adopt a local independent control architecture, and the control logic is mainly based on threshold setting and manual experience. However, the existing gate control system still has the following problems: (1) There is a lack of linkage and coordination mechanism between distributed systems; (2) The system response is delayed and the emergency response capability is weak; (3) The scheduling algorithm is single and lacks an intelligent prediction mechanism; (4) The control target is single and lacks multi-objective scheduling optimization; (5) There is no state feedback mechanism and the control closed-loop capability is insufficient.

[0003] In summary, there is an urgent need for a multi-gate collaborative control system with intelligent prediction, flexible scheduling, linkage coordination and state feedback capabilities to realize intelligent modeling, prediction drive, closed-loop scheduling and linkage execution of multi-gate collaborative control systems under complex hydrological environments. Summary of the Invention

[0004] The present invention provides a multi-gate coordinated control system and a control method thereof to solve the technical problems of traditional gate control systems, such as lack of coordinated scheduling mechanism, insufficient prediction capability, delayed control response and rigid regulation strategy.

[0005] The multi-gate coordinated control system and control method of the present invention specifically include the following technical solutions:

[0006] A multi-gate coordinated control method comprises the following steps:

[0007] S1. Real-time collection of multi-gate hydrological status data, and performing numerical validity checks and logical judgment on the multi-gate hydrological status data. Preprocessing the logically judged multi-gate hydrological status data to obtain structured multi-gate hydrological status data.

[0008] S2. Based on structured multi-gate hydrological status data, a graph neural network algorithm optimized with a graph attention mechanism is used to predict the future hydrological status of each gate node, resulting in a structured hydrological status prediction result.

[0009] S3. Based on the structured hydrological state prediction results, a state scoring model is constructed to generate a water level target score, a fluctuation stability score, and an operation energy consumption score as scoring indicators. The water level target score, fluctuation stability score, and operation energy consumption score are weighted to obtain a comprehensive score. Based on the comprehensive score, an optimal multi-gate coordinated opening strategy is generated to control gate operation and collect multi-gate hydrological state data in real time to achieve a closed-loop system.

[0010] Preferably, the S1 specifically includes:

[0011] Multi-gate hydrological status data includes hydrological data and equipment status data, where hydrological data includes upstream water level, downstream water level and flow rate; equipment status data includes motor operating current and gate opening.

[0012] Preferably, the S1 specifically includes:

[0013] Perform numerical validity check on the multi-gate hydrological status data to check whether the multi-gate hydrological status data is within the set range; perform logical judgment based on the multi-gate hydrological status data after numerical validity check. When the logic is normal, the multi-gate collaborative control system will maintain the current stable operating state; when the logic is abnormal, the local emergency control logic will be triggered to execute a rapid control response.

[0014] Preferably, the S2 specifically includes:

[0015] In the graph neural network algorithm optimized based on the graph attention mechanism, the multi-gate collaborative control system is constructed as a dynamic directed graph, the structured multi-gate hydrological status data is split according to the gate nodes, and the feature vectors of the gate nodes are generated in a standardized manner.

[0016] Preferably, the S2 specifically includes:

[0017] Based on the feature vector of the gate node, an edge weight attention mechanism is proposed on the basis of the graph neural network algorithm, the edge attention weight is defined, and the influence intensity between the gate nodes is adjusted according to the real-time hydrological status.

[0018] Preferably, the S2 specifically includes:

[0019] Based on the edge weight attention mechanism, a time series attention mechanism is proposed to obtain the prediction features of the gate node at the next moment; the prediction features are input into the long short-term memory neural network for feature recognition, and the predicted value of the hydrological variable of the target gate node is output as the structured hydrological state prediction result.

[0020] Preferably, the S3 specifically includes:

[0021] A water level target score is generated based on the relative error between the upstream water level of the gate node and the set target water level value; a fluctuation stability score is generated based on the difference in upstream water levels of the gate node at different times, which is used to measure the stability of the water level prediction sequence; and an action energy consumption score is generated based on the difference in gate openings of the gate node at different times.

[0022] Preferably, the S3 specifically includes:

[0023] Construct alternative opening strategy combinations, calculate the comprehensive score of each opening strategy combination in the future prediction time window, and select the opening strategy combination with the highest comprehensive score as the multi-gate coordinated optimal opening strategy for the current period.

[0024] A multi-gate coordinated control system includes the following parts:

[0025] Gate execution and perception module, edge control processing module, data processing module, graph modeling and prediction module, and strategy calculation module;

[0026] The gate execution and perception module is used to control gate movement and collect multi-gate hydrological status data in real time, and transmit the multi-gate hydrological status data to the edge control processing module; and receives the multi-gate coordinated optimal opening strategy issued by the edge control processing module to drive the gate actuator in the gate execution and perception module to control gate movement;

[0027] The edge control processing module is used to perform emergency control of the gates, and simultaneously complete the validity detection and logical judgment of the multi-gate hydrological status data, obtain the processed multi-gate hydrological status data, and upload the processed multi-gate hydrological status data to the data processing module; at the same time, it receives the multi-gate coordinated optimal opening strategy transmitted by the strategy calculation module, and transmits the multi-gate coordinated optimal opening strategy to the gate execution and perception module;

[0028] The data processing module is used to perform structured processing on the multi-gate hydrological status data processed by the edge control processing module to form structured multi-gate hydrological status data, and pass the structured multi-gate hydrological status data to the graph modeling prediction module;

[0029] The graph modeling and prediction module obtains the structured hydrological state prediction results through the graph neural network algorithm optimized based on the graph attention mechanism, and outputs the structured hydrological state prediction results to the strategy calculation module;

[0030] The strategy calculation module is used to generate the multi-gate coordinated optimal opening strategy through the state scoring model based on the structured hydrological state prediction results of the graph modeling prediction module, and send the multi-gate coordinated optimal opening strategy to the edge control processing module.

[0031] The beneficial effects of the technical solution of the present invention are:

[0032] 1. Deploying preliminary logic judgment and emergency control functions in the edge control processing module can independently complete closed-loop control under network interruptions or emergency conditions, avoiding scheduling risks caused by cloud delays or interruptions, and significantly improving the stability and safety of the multi-gate collaborative control system.

[0033] 2. A dynamic directed graph structure is constructed with gates as nodes and hydraulic regulation effects as edges. The edge weight attention mechanism and time series attention mechanism are introduced. A graph neural network algorithm optimized based on the graph attention mechanism is proposed, which significantly enhances the trend modeling and prediction accuracy of the hydrological status of the multi-gate collaborative control system.

[0034] 3. Through the state scoring model, including water level target scoring, fluctuation stability scoring and action energy consumption scoring, the scheduling strategy screening and optimization driven by prediction is realized, the transparency, debuggability and target adaptability of the scheduling strategy are improved, and the multi-objective integrated control of flood control, ecology, energy conservation, etc. is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a structural diagram of a multi-gate coordinated control system according to the present invention;

[0036] Figure 2 This is a flow chart of a multi-gate coordinated control method described in the present invention. DETAILED DESCRIPTION

[0037] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0038] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0039] The specific scheme of a multi-gate coordinated control system and a control method thereof provided by the present invention is described in detail below with reference to the accompanying drawings.

[0040] Refer to the attached Figure 1 , which shows a structural diagram of a multi-gate coordinated control system provided by an embodiment of the present invention, the system includes the following parts:

[0041] Gate execution and perception module, edge control processing module, data processing module, graph modeling and prediction module, and strategy calculation module;

[0042] The gate execution and perception module is used to control gate movement and collect multi-gate hydrological status data in real time, and transmit the collected multi-gate hydrological status data to the edge control processing module; and receives the multi-gate coordinated optimal opening strategy issued by the edge control processing module to drive the gate actuator in the gate execution and perception module to control gate movement;

[0043] The edge control processing module is used to perform emergency control of the gates in the event of network interruption or emergency. It also completes the validity detection and logical judgment of the multi-gate hydrological status data, obtains the processed multi-gate hydrological status data, and uploads the processed multi-gate hydrological status data to the data processing module. At the same time, it receives the multi-gate coordinated optimal opening strategy transmitted by the strategy calculation module and transmits the multi-gate coordinated optimal opening strategy to the gate execution and perception module.

[0044] The data processing module is used to perform structured processing on the multi-gate hydrological status data processed by the edge control processing module to form structured multi-gate hydrological status data, and pass the structured multi-gate hydrological status data to the graph modeling prediction module;

[0045] The graph modeling and prediction module obtains the structured hydrological state prediction results through the graph neural network algorithm optimized based on the graph attention mechanism, and outputs the structured hydrological state prediction results to the strategy calculation module;

[0046] The strategy calculation module is used to generate the multi-gate coordinated optimal opening strategy through the state scoring model based on the structured hydrological state prediction results of the graph modeling prediction module, and send the multi-gate coordinated optimal opening strategy to the edge control processing module.

[0047] Refer to the attached Figure 2 , which shows a flow chart of a multi-gate coordinated control method provided by an embodiment of the present invention, the method comprising the following steps:

[0048] S1. Real-time collection of multi-gate hydrological status data, and numerical validity detection and logical judgment of the multi-gate hydrological status data; pre-processing of the multi-gate hydrological status data after logical judgment to obtain structured multi-gate hydrological status data.

[0049] S101. Real-time multi-gate hydrological status data is collected through sensors and other equipment deployed at each gate. Sensors and other equipment include ultrasonic water level sensors, current sensors, electromagnetic flowmeters, and position encoders. Multi-gate hydrological status data includes hydrological data and equipment status data. Hydrological data includes upstream and downstream water levels (in meters (m) and flow velocity (in meters per second (m / s)). Equipment status data includes motor operating current (in amperes (A)) and gate opening (in percentage (%)).

[0050] After the gate execution and perception module uploads the collected multi-gate hydrological status data to the edge control processing module, the edge control processing module first performs a numerical validity check on the multi-gate hydrological status data. This check verifies whether the multi-gate hydrological status data is within a set range, which is determined based on expert experience. For example, the validity range for the upstream water level designed for the multi-gate coordinated control system is 1.0 to 9.5 meters, and the validity range for flow velocity is 0.1 to 4.8 meters per second. The module then uses a sliding average and range judgment method to remove noise. The edge control processing module then performs a logical judgment based on the multi-gate hydrological status data after the numerical validity check. If the logic is normal, the multi-gate coordinated control system will maintain its current stable operating state. The edge control processing module only uploads the multi-gate hydrological status data after the numerical validity check to the data processing module at a reasonable interval. The period setting depends on the frequency of hydrological status changes, the real-time requirements of the multi-gate coordinated control system, and the bandwidth of the communication system. It can be set to a reasonable interval of 5, 10, or 30 minutes. If the logic is abnormal, the local emergency control logic in the edge control processing module is triggered to implement a rapid control response. Logical judgment is based on expert experience. For example, experts may set a water level-flow rate relationship model based on their experience. If the flow rate does not match the water level change, it is considered a logical abnormality.

[0051] Finally, the multi-gate hydrological status data processed by the edge control processing module is uploaded to the data processing module.

[0052] The local emergency control logic under logical anomalies specifically includes but is not limited to: ① If the upstream and downstream water levels of a gate change by more than 0.5 meters within 5 minutes, the multi-gate collaborative control system will automatically execute multiple response strategies such as status marking, starting pre-discharge mode, triggering early warning, and limiting the subsequent opening adjustment rhythm; ② If the motor continues to run for a timeout or the gate is stuck, it enters the "equipment protection mode" and sends a text message / sound and light alarm prompt; ③ If the upstream water level is higher than the set flood discharge warning line, such as the set flood discharge warning line is 5.20 meters, the "safety pre-discharge strategy" will be activated to automatically adjust the gate opening to the preset safety opening (such as 60%) to lower the upstream water level.

[0053] S102. After receiving the multi-gate hydrological status data processed by the edge control processing module, the data processing module performs preprocessing operations on the multi-gate hydrological status data processed by the edge control processing module to form structured multi-gate hydrological status data; the preprocessing operations include time alignment, cleaning, structuring, standardization, etc., which are all technical means well known to technical personnel in this field; finally, the structured multi-gate hydrological status data is sent to the graph modeling prediction module.

[0054] S2. Based on the structured multi-gate hydrological status data, the future hydrological status of each gate node is predicted through the graph neural network algorithm optimized based on the graph attention mechanism, and the structured hydrological status prediction results are obtained.

[0055] S201. Based on structured multi-gate hydrological status data, a graph neural network algorithm based on graph attention mechanism optimization is proposed to construct the entire multi-gate collaborative control system as a dynamic directed graph. ,in represents a set of gate nodes, Indicates the Gate nodes, Indicates the number of gate nodes; edge set It consists of pairs of nodes that have actual hydraulic control influence relationships between gates. Represents a gate node Gate Node The actual impact direction on hydraulic regulation. Gate node Gate node The neighbor gate node.

[0056] The structured multi-gate hydrological status data is split according to the gate nodes and standardized to generate gate nodes The eigenvector of , the eigenvector include: Upstream water level at any moment ; Downstream water level at the moment ; Gate opening at all times ; Time flow rate ; Motor operating current at all times .

[0057] S202. The influence between gates is not static, but is affected by the current water level difference, water flow changes, gate position, etc. Therefore, an improved method of edge weight attention mechanism is proposed based on the graph neural network algorithm, and the edge attention weight is defined as:

[0058] ,

[0059] Among them, the edge attention weight Indicates Time, gate node Gate Node The impact weight of is a normalization function used to normalize the attention score of the neighbor gate node into a weight distribution; Represents the attention score weight vector, obtained from the experiment, used to combine the concatenated features Perform linear scoring, the higher the score, the higher the gate node Pay more attention to gate nodes ; represents the attention score weight vector The transpose of Represents an activation function, such as ReLU or LeakyReLU, used to linearly combine The results are nonlinearly mapped to enhance the expressive power of the edge weight attention mechanism; Represents the edge feature mapping weight matrix, obtained from experiments, used to perform linear transformation on the spliced ​​node and edge features; Represents a gate node exist The eigenvector of the moment; Represents a gate node exist The eigenvector of the moment; ,express The hydraulic transmission intensity at the moment; Represents a vector concatenation operation.

[0060] By introducing the edge-weighted attention mechanism, the graph neural network algorithm optimized based on the graph attention mechanism can adaptively adjust the influence intensity between each gate node according to the real-time hydrological status, thereby more accurately characterizing the hydraulic conduction relationship between gates and improving the prediction accuracy and generalization ability under complex hydrological conditions (such as sudden floods and asynchronous regulation).

[0061] S203. Since the hydrological evolution of the multi-gate collaborative control system also has time correlation and time series accumulation effects, a time series attention mechanism is further proposed based on the edge weight attention mechanism.

[0062] Define gate nodes The next moment, The prediction features at the moment are:

[0063] ,

[0064] in, Represents a gate node exist Predictive feature representation of the moment; Represents a gate node Neighbor gate node A collection of Represents the activation function, such as ReLU or LeakyReLU, responsible for the aggregation result Perform nonlinear processing to enhance the node feature expression ability; edge attention weight Indicates Time, gate node Gate Node The impact weight of Represents the feature map weight matrix for temporal state evolution, used to transform neighbor gate nodes exist The eigenvector at time Projection into a mapping space, obtained by experiment; Represents a gate node exist The feature vector of the moment.

[0065] By introducing a time series attention mechanism, we can capture the dominant role of historical multi-gate hydrological status data at critical moments in the evolution trend of the current target gate node, thereby improving the prediction accuracy and robustness of the multi-gate coordinated control system in complex time series environments. The historical multi-gate hydrological status data is obtained from a database.

[0066] S204, will The prediction features are input into the existing long short-term memory neural network for feature recognition. The long short-term memory neural network finally outputs the predicted value of the hydrological variable of the identified target gate node. The predicted value of the hydrological variable of the target gate node includes the gate node exist Upstream water level at the moment , downstream water level , gate opening and flow rate , which is output as the structured hydrological state prediction result. The existing long-short-term memory neural network technology is relatively mature and can be directly adopted.

[0067] S3. Based on the structured hydrological state prediction results, a state scoring model is constructed to generate a water level target score, a fluctuation stability score, and an action energy consumption score as scoring indicators; the water level target score, the fluctuation stability score, and the action energy consumption score are weighted to obtain a comprehensive score; based on the comprehensive score, a multi-gate coordinated optimal opening strategy is generated to control gate action, and multi-gate hydrological state data is collected in real time to achieve a closed loop.

[0068] S301, based on the structured hydrological state prediction results, the strategy calculation module calculates the gate node A state scoring model is constructed to generate three scoring indicators, including water level target score, fluctuation stability score, and action energy consumption score. The specific formula is as follows:

[0069] ,

[0070] ,

[0071] ,

[0072] in, Indicates the water level target score, used to calculate the gate node exist Upstream water level at the moment and target water level The relative error between them is as follows: the smaller the error, the higher the water level target score, and the score approaches 1; Γ is the number of prediction time steps; Gate node exist Upstream water level at all times; It is the target water level value set based on expert experience;

[0073] It represents the fluctuation stability score, which is used to measure the stability of the water level forecast series. If the water level changes slowly in the future, the higher the fluctuation stability score is, the closer the score is to 1. Gate node exist Upstream water level at all times; Gate node exist Upstream water level at all times;

[0074] Indicates the energy consumption score of the action, used to calculate the gate node exist Gate opening at the time and Gate opening at all times The smaller the difference, the higher the energy consumption score, and the score approaches 1; Indicates the maximum allowable range of gate opening, used for normalization. When set to 100, it means the adjustment range is calculated as a percentage.

[0075] S302: Perform weighted calculation on the water level target score, fluctuation stability score, and action energy consumption score to obtain a comprehensive score. :

[0076] ,

[0077] in, 、 、 is the weight coefficient, which is set according to the expert experience method to further construct several alternative opening strategy combinations, such as: flood control priority strategy combination , energy-saving priority strategy combination .

[0078] Calculate the comprehensive score of each opening strategy combination in the future prediction time window, and select the opening strategy combination with the highest comprehensive score as the multi-gate coordinated optimal opening strategy for the current period.

[0079] By introducing a state scoring model, comprehensive scoring and strategy screening of the future behavior of multiple gates can be achieved, taking into account water level control, stability and operating costs, and realizing a multi-gate collaborative scheduling execution process that is interpretable, flexible and engineering feasible.

[0080] S303. The strategy calculation module sends the multi-gate coordinated optimal opening strategy to the edge control processing module. After receiving the multi-gate coordinated optimal opening strategy, the edge control processing module drives the gate actuator in the gate execution and perception module to control the gate action according to the specified opening, and collects the hydrological status data of multiple gates in real time, and uploads it to the edge control processing module to achieve a closed loop.

[0081] In summary, a multi-gate collaborative control system and a control method thereof are completed.

[0082] The order in which the embodiments of the invention are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0083] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0084] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A multi-gate coordinated control method, characterized in that: The following steps are involved: S1. Real-time collection of multi-gate hydrological status data and numerical validity testing of the data to check whether it is within the set range; Based on the multi-gate hydrological status data after numerical validity detection, logical judgment is performed. When the logic is normal, the multi-gate coordinated control system will maintain the current stable operating state; when the logic is abnormal, the local emergency control logic will be triggered to perform a rapid control response. The multi-gate hydrological status data after logical judgment is preprocessed to obtain structured multi-gate hydrological status data. S2. Based on structured multi-gate hydrological status data, a graph neural network algorithm optimized with a graph attention mechanism is designed. The multi-gate coordinated control system is constructed as a dynamic directed graph. The edge set in the dynamic directed graph consists of node pairs with actual hydraulic control influence relationships between gates. The influence between gates is not static. Therefore, based on the graph neural network algorithm, an improved edge weighted attention mechanism is proposed to adaptively adjust the influence strength between gate nodes based on real-time hydrological status. Based on the edge weighted attention mechanism, a time series attention mechanism is proposed to obtain the predicted features of the gate nodes at the next moment. The predicted features are input into a long short-term memory neural network for feature recognition, and the predicted value of the hydrological variable of the target gate node is output as the structured hydrological status prediction result. S3. Based on the structured hydrological state prediction results, a state scoring model is constructed to generate a water level target score, a fluctuation stability score, and an operation energy consumption score as scoring indicators. The water level target score, fluctuation stability score, and operation energy consumption score are weighted to obtain a comprehensive score. Based on the comprehensive score, an optimal multi-gate coordinated opening strategy is generated to control gate operation and collect multi-gate hydrological state data in real time to achieve a closed-loop system.

2. The multi-gate coordinated control method according to claim 1, characterized in that: Said S1 specifically includes: Multi-gate hydrological status data includes hydrological data and equipment status data, where the hydrological data includes upstream water level, downstream water level and flow rate; the equipment status data includes motor operating current and gate opening.

3. The multi-gate coordinated control method according to claim 1, characterized in that: Said S2 specifically includes: In the graph neural network algorithm optimized based on the graph attention mechanism, the structured multi-gate hydrological status data is split according to the gate nodes, and the feature vectors of the gate nodes are generated by standardization.

4. The multi-gate coordinated control method according to claim 3, characterized in that: Said S2 specifically includes: The edge-weighted attention mechanism defines the edge attention weight based on the feature vector of the gate node, the feature vector of the neighboring gate nodes, and the hydraulic transmission strength.

5. The multi-gate coordinated control method according to claim 1, characterized in that: Said S3 specifically includes: A water level target score is generated based on the relative error between the upstream water level of the gate node and the set target water level value; a fluctuation stability score is generated based on the difference in upstream water levels of the gate node at different times, which is used to measure the stability of the water level prediction sequence; and an action energy consumption score is generated based on the difference in gate openings of the gate node at different times.

6. The multi-gate coordinated control method according to claim 1, characterized in that: Said S3 specifically includes: Construct alternative opening strategy combinations, calculate the comprehensive score of each opening strategy combination in the future prediction time window, and select the opening strategy combination with the highest comprehensive score as the multi-gate coordinated optimal opening strategy for the current period.

7. A multi-gate coordinated control system, applied to the multi-gate coordinated control method according to claim 1, characterized in that: Includes the following sections: Gate execution and perception module, edge control processing module, data processing module, graph modeling and prediction module, and strategy calculation module; The gate execution and perception module is used to control gate movement and collect multi-gate hydrological status data in real time, and transmit the multi-gate hydrological status data to the edge control processing module; and receives the multi-gate coordinated optimal opening strategy issued by the edge control processing module to drive the gate actuator in the gate execution and perception module to control gate movement; The edge control processing module is used to perform emergency control of the gates, and simultaneously complete the validity detection and logical judgment of the multi-gate hydrological status data, obtain the processed multi-gate hydrological status data, and upload the processed multi-gate hydrological status data to the data processing module; at the same time, it receives the multi-gate coordinated optimal opening strategy transmitted by the strategy calculation module, and transmits the multi-gate coordinated optimal opening strategy to the gate execution and perception module; The data processing module is used to perform structured processing on the multi-gate hydrological status data processed by the edge control processing module to form structured multi-gate hydrological status data, and pass the structured multi-gate hydrological status data to the graph modeling prediction module; The graph modeling and prediction module obtains the structured hydrological state prediction results through the graph neural network algorithm optimized based on the graph attention mechanism, and outputs the structured hydrological state prediction results to the strategy calculation module; The strategy calculation module is used to generate the multi-gate coordinated optimal opening strategy through the state scoring model based on the structured hydrological state prediction results of the graph modeling prediction module, and send the multi-gate coordinated optimal opening strategy to the edge control processing module.

Images