A water conservancy project scheduling system based on artificial intelligence

Abstract

The present invention discloses an artificial intelligence-based water conservancy project scheduling system, comprising a water conservancy data monitoring module, a water level prediction module, a water conservancy scheduling module and a water conservancy power generation module; the water conservancy data monitoring module is used to monitor and obtain water conservancy data in real time, and to obtain meteorological data in real time; the water level prediction module is used to obtain historical water conservancy data and historical meteorological data, establish a BP neural network model, and obtain predicted water level data; the water conservancy scheduling module is used to obtain a first water conservancy scheduling strategy based on the predicted water level data using a particle swarm algorithm, and establish a fluid simulation model based on the water conservancy scheduling strategy to obtain a second water conservancy scheduling strategy; the water conservancy power generation module is used to determine the water conservancy power generation strategy using edge computing, realize intelligent water conservancy project scheduling, ensure the feasibility of the water conservancy scheduling strategy and the reliability of actual scheduling, and supply power to power users according to the water conservancy power generation strategy, which is conducive to making full use of hydropower generation and improving energy utilization.

Description

Technical Field

[0001] The present invention relates to a water conservancy project dispatching system based on artificial intelligence. Background Art

[0002] In recent years, as water conservancy project scheduling is a key link in ensuring the efficient use of water resources, traditional scheduling methods are difficult to cope with complex and changeable hydrological conditions and multi-objective demands. With the development of artificial intelligence technology, applying it to water conservancy project scheduling has become an important way to improve scheduling efficiency.

[0003] At present, the Chinese invention with publication number CN119151238A discloses a real-time scheduling method for water conservancy engineering systems based on neural network technology. Although a deep learning model is used to predict the future water level of the reservoir and the vulnerability coefficient is calculated based on flood control data; the flood control coefficient is calculated based on the rainfall coefficient, rain cloud influence coefficient and vulnerability coefficient, and a flood control regulation model is constructed, which effectively improves the flexibility of real-time scheduling of water conservancy engineering systems, there is no detailed regulation of the future water level of the reservoir, and there is no effective utilization of the water energy generated after the water conservancy regulation, which is not conducive to the full implementation of new energy power generation. Summary of the Invention

[0004] The technical problem solved by the present invention is that the existing technology does not carry out detailed regulation of the future water level of the reservoir, and does not effectively utilize the water energy generated after water conservancy regulation, resulting in energy waste and being unfavorable for fully carrying out new energy power generation.

[0005] In order to solve the above technical problems, the present invention provides a water conservancy project scheduling system based on artificial intelligence, which is characterized by comprising a water conservancy data monitoring module, a water level prediction module, a water conservancy scheduling module and a water conservancy power generation module;

[0006] The water conservancy data monitoring module is used to monitor and obtain water conservancy data in real time using sensors, and the meteorological monitoring platform is used to obtain meteorological data in real time;

[0007] The water level prediction module is used to obtain historical water conservancy data and historical meteorological data, and establish a BP neural network model based on the historical water conservancy data and historical meteorological data, and input the water conservancy data and the meteorological data into the BP neural network model to obtain predicted water level data;

[0008] The water conservancy scheduling module is used to obtain a first water conservancy scheduling strategy using a particle swarm algorithm based on the predicted water level data, establish a fluid simulation model based on the water conservancy scheduling strategy and obtain simulation results, and adjust the first water conservancy scheduling strategy based on the simulation results to obtain a second water conservancy scheduling strategy;

[0009] The hydropower generation module is used to obtain user demand information, use edge computing to obtain power supply users, and determine a hydropower generation strategy based on the user demand information and the second water conservancy scheduling strategy, and perform hydropower supply for the power supply users through the hydropower generation strategy;

[0010] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0011] The water conservancy data monitoring module includes a water conservancy monitoring unit and a meteorological monitoring module;

[0012] The water conservancy monitoring unit is used to obtain water conservancy data in real time using sensors, and the water conservancy data includes water level data, water flow data, evaporation data, water quality data and gate data;

[0013] The meteorological monitoring module is used to obtain meteorological data within a future time period from the meteorological monitoring platform in real time, and the meteorological data includes temperature, humidity, air pressure, wind speed, wind direction, rainfall time, rainfall intensity and rainfall speed;

[0014] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0015] The water level prediction module includes a prediction model unit and a water level prediction unit;

[0016] The prediction model unit is used to obtain historical water conservancy data and historical meteorological data, input the historical water conservancy data and the historical meteorological data into the BP neural network for training, and use the historical water conservancy data and the historical meteorological data as input and the historical water level data as output to establish a BP neural network model;

[0017] The water level prediction unit is used to input the water conservancy data and the meteorological data of each water station into a BP neural network model, obtain the predicted water level data of each station, retrieve the predicted water level of the water station in the water level data, define the water station whose predicted water level is greater than or equal to a first expected threshold as a flood warning water station, and define the water station whose predicted water level is less than or equal to a second expected threshold as a drought warning water station;

[0018] The historical water conservancy data includes historical water level data, historical water flow data, historical evaporation data, historical water quality data and historical gate data;

[0019] The historical meteorological data includes historical temperature, historical humidity, historical air pressure, historical wind speed, historical wind direction, historical rainfall time, historical rainfall intensity and historical rainfall speed;

[0020] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0021] The water conservancy scheduling module includes a scheduling strategy unit and a simulation unit;

[0022] The scheduling strategy unit is used to obtain flood particle information based on the flood warning water station, the flood particle information including the water station type, water station location data, water station water level data and water station sluice data of other water stations except the flood warning water station, obtain the flood water station location data of the flood warning water station, obtain the flood water station scheduling distance based on the water station location data and the flood water station location data, determine the flood resistance particle value of each water station based on the flood water station scheduling distance and the water station type, obtain the water station with the largest flood resistance particle value, and define it as the flood discharge station of the flood warning water station;

[0023] Obtain drought particle information based on drought warning water stations, the drought particle information including types of water stations other than the drought warning water stations, water station location data, water level data of the water stations, and water gate data of the water stations; obtain drought water station location data of the drought warning water stations; obtain drought water station dispatching distances based on the water station location data and the drought water station location data; determine drought resistance particle values for each water station based on the drought water station dispatching distances and the water station types; obtain the water station with the largest drought resistance particle value and define it as the drought resistance water station of the drought warning water station; repeat the above steps to obtain a first water conservancy dispatching strategy;

[0024] The simulation unit is used to establish a fluid simulation model according to the water conservancy data, determine the first fluid incremental data according to the predicted water level data, and update the fluid simulation model according to the first fluid incremental data, obtain the second fluid adjustment data through the first water conservancy scheduling strategy, and input the second fluid adjustment data into the fluid simulation model to obtain the fluid simulation data, retrieve the simulated flow of the fluid simulation data, define the flow of the simulated flow greater than or equal to the first expected threshold as a first abnormal flow, define the flow of the fluid simulation model less than or equal to the second expected threshold as a second abnormal flow, define the first water conservancy scheduling strategy corresponding to the first abnormal flow or the second abnormal flow as an abnormal fluid strategy, and adjust the abnormal fluid strategy according to the first abnormal flow and the second abnormal flow to obtain the second water conservancy scheduling strategy;

[0025] Determining the drought-resistant particle value of each water station according to the drought water station scheduling distance and the water station type includes:

[0026] Obtain a dispatching distance coefficient based on the dispatching distance of the drought water station, and obtain a dispatching water station coefficient based on the type of the water station. The dispatching water station coefficient of the flood warning water station is 1, the dispatching water station coefficient of the normal operating water station is 0, and the dispatching water station coefficient of the drought warning water station is -1. Perform a weighted calculation on the dispatching distance coefficient and the dispatching water station coefficient to obtain a drought-resistant particle value.

[0027] Determining the flood-fighting particle value of each water station according to the flood water station dispatching distance and the water station type includes:

[0028] Obtain a dispatching distance coefficient according to the flood water station dispatching distance, and obtain a dispatching water station coefficient according to the water station type. The dispatching water station coefficient of the flood warning water station is -1, the dispatching water station coefficient of the normal operating water station is 0, and the dispatching water station coefficient of the drought warning water station is 1. Perform a weighted calculation on the dispatching distance coefficient and the dispatching water station coefficient to obtain a post-flood resistance particle value.

[0029] The types of water stations include flood warning water stations, drought warning water stations and normal operating water stations;

[0030] The fluid simulation data includes simulated water level, simulated flow rate and simulated flow velocity;

[0031] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0032] The hydropower generation module includes a power generation information unit and a power generation strategy unit;

[0033] The power generation information unit is used to obtain user demand information of the water station, and determine the user power supply distance according to the location of the water station and the user, and use edge computing to determine the power supply user according to the user power supply distance;

[0034] The power generation strategy unit is used to obtain the scheduling water station pair information according to the second water conservancy scheduling strategy, call the gate data of the scheduling water station pair information, and obtain the water head through the gate data, and use the water conservancy power generation calculation formula to obtain the single-gate power supply, double-gate power supply, and triple-gate power supply of the water station through the water head... The peak power consumption per unit time is obtained through the user demand information of the power supply user, and the water conservancy power generation strategy is determined according to the peak power consumption. When the peak power consumption is less than or equal to the single-gate power supply, only the single gate is opened. When the peak power consumption is greater than the single-gate power supply and less than or equal to the double-gate power supply, the double gates are opened. When the peak power consumption is greater than or equal to the double-gate power supply and less than or equal to the triple-gate power supply, the three gates are opened. Repeat the above steps to obtain the water conservancy power generation strategy, and dispatch the water station gates according to the water conservancy power generation strategy;

[0035] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0036] The establishment of the BP neural network model specifically includes:

[0037] Input the historical water conservancy data and the historical meteorological data into the input layer of the BP neural network, obtain a prediction factor based on the historical water conservancy data and the historical meteorological data, randomly obtain a first weight value and a first bias value of the prediction factor, and use a first activation function to obtain a first prediction value based on the first weight value and the first bias value, use the first prediction value as the input of the hidden layer, and use a loss function to obtain a loss value based on the first prediction value and the true value, adjust the first weight value and the first bias value based on the loss value, obtain a second weight value and a second bias value, repeat the above steps n times until the loss value is less than or equal to the loss expectation threshold, then obtain a final weight value and a final bias value, and use a second activation function based on the final weight value and the final bias value to obtain predicted water level data as the output of the output layer;

[0038] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0039] The water level data includes the average water level per unit time and the water level distribution per unit time;

[0040] The water flow data includes flow rate per unit time, flow velocity per unit time and runoff per unit time;

[0041] The evaporation data includes the average evaporation per unit time;

[0042] The water quality data include water temperature, turbidity, dissolved oxygen content, pH value, chemical oxygen demand, total phosphorus and total nitrogen;

[0043] The gate data includes the water level before the gate, the water level after the gate, the gate opening, the gate operating condition and the flow rate through the gate;

[0044] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0045] The hydropower generation calculation formula is as follows:

[0046] ;

[0047] Among them, Q is the hydropower generation, q is the flow rate per unit time, H is the head value, g is the acceleration of gravity, and t is the power generation time;

[0048] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0049] Determining the power supply user according to the power supply distance of the user by using edge computing includes:

[0050] Arrange the power supply distances of the users in ascending order, select the top m users, and determine them as power supply users;

[0051] As a preferred solution of the artificial intelligence-based water conservancy project scheduling system of the present invention, wherein:

[0052] The dispatching water station pair information includes the first water station data, the second water station data and the gate data;

[0053] The user demand information includes the power consumption area number, power demand and power consumption time.

[0054] The beneficial effects of the present invention are as follows: the present invention uses a BP neural network model to predict the water level of a water station based on historical water conservancy data and historical meteorological data, and obtains a flood warning water station and a drought warning water station based on the prediction results. The particle swarm algorithm is used to obtain a first water conservancy scheduling strategy based on the flood warning water station and the drought warning water station, and flood water is diverted from the flood warning water station to the drought warning water station, thereby solving flood problems and drought problems at the same time, making full use of water resources, and realizing intelligent water conservancy project scheduling.

[0055] The fluid simulation model is used to simulate the scheduling process according to the first water conservancy scheduling strategy, obtain the scheduling results, and obtain the second water conservancy scheduling strategy based on the scheduling results to improve the shortcomings of the first water conservancy scheduling strategy and ensure the feasibility of the water conservancy scheduling strategy and the safety and reliability of actual scheduling.

[0056] Edge computing is used to obtain power supply users, and the hydropower generation strategy is determined based on user demand information and the second water conservancy scheduling strategy. Power is generated according to the hydropower generation strategy and power is supplied to power users, which is conducive to making full use of hydropower generation and improving energy utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic diagram of the basic flow of a water conservancy project scheduling system based on artificial intelligence is provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0059] Example, see Figure 1 , as an embodiment of the present invention, provides a water conservancy project scheduling system based on artificial intelligence, including a water conservancy data monitoring module, a water level prediction module, a water conservancy scheduling module and a water conservancy power generation module;

[0060] The water conservancy data monitoring module is used to monitor and obtain water conservancy data in real time using sensors, and the meteorological monitoring platform is used to obtain meteorological data in real time;

[0061] The water level prediction module is used to obtain historical water conservancy data and historical meteorological data, and establish a BP neural network model based on the historical water conservancy data and historical meteorological data, and input the water conservancy data and the meteorological data into the BP neural network model to obtain predicted water level data;

[0062] The water conservancy scheduling module is used to obtain a first water conservancy scheduling strategy using a particle swarm algorithm based on the predicted water level data, establish a fluid simulation model based on the water conservancy scheduling strategy and obtain simulation results, and adjust the first water conservancy scheduling strategy based on the simulation results to obtain a second water conservancy scheduling strategy;

[0063] The hydropower generation module is used to obtain user demand information, use edge computing to obtain power supply users, and determine a hydropower generation strategy based on the user demand information and the second hydropower scheduling strategy, and provide hydropower supply to the power supply users through the hydropower generation strategy.

[0064] In this embodiment, the BP neural network model is a multi-layer feedforward neural network trained according to the error back propagation algorithm;

[0065] In this embodiment, the particle swarm algorithm is a computer algorithm that is based on the observation of animal group activity behavior and utilizes the information sharing of individuals in the group to make the movement of the entire group evolve from disorder to order in the problem-solving space, thereby obtaining the optimal solution;

[0066] In this embodiment, the fluid simulation model utilizes computer technology and numerical calculation to simulate the actual fluid flow situation;

[0067] In this embodiment, edge computing provides cloud services and IT environment services to application developers and service providers at the edge of the network, with the goal of providing computing, storage, and network bandwidth close to data input or users.

[0068] In this embodiment, the water level of the water station is predicted using the BP neural network model based on historical water conservancy data and historical meteorological data, and flood warning water stations and drought warning water stations are obtained based on the prediction results. The particle swarm algorithm is used to obtain the first water conservancy scheduling strategy based on the flood warning water stations and drought warning water stations, and flood water is diverted from the flood warning water station to the drought warning water station, thereby solving the flood and drought problems at the same time, making full use of water resources, and realizing intelligent water conservancy project scheduling.

[0069] The fluid simulation model is used to simulate the scheduling process according to the first water conservancy scheduling strategy, obtain the scheduling results, and obtain the second water conservancy scheduling strategy based on the scheduling results to improve the shortcomings of the first water conservancy scheduling strategy and ensure the feasibility of the water conservancy scheduling strategy and the safety and reliability of actual scheduling.

[0070] Edge computing is used to obtain power supply users, and the hydropower generation strategy is determined based on user demand information and the second water conservancy scheduling strategy. Power is generated according to the hydropower generation strategy and power is supplied to power users, making full use of hydropower generation and improving energy utilization.

[0071] The water conservancy data monitoring module includes a water conservancy monitoring unit and a meteorological monitoring module;

[0072] The water conservancy monitoring unit is used to obtain water conservancy data in real time using sensors, and the water conservancy data includes water level data, water flow data, evaporation data, water quality data and gate data;

[0073] The meteorological monitoring module is used to obtain meteorological data within a future time period from the meteorological monitoring platform in real time. The meteorological data includes temperature, humidity, air pressure, wind speed, wind direction, rainfall time, rainfall intensity and rainfall speed.

[0074] In this embodiment, a water level sensor is used to obtain water level data, a flow rate sensor is used to obtain water flow data, an evaporation sensor is used to obtain evaporation data, and a water quality sensor is used to obtain water quality data;

[0075] In this embodiment, real-time monitoring of water conservancy data and meteorological data provides detailed and reliable data support for establishing a BP neural network model to predict water levels at water stations, and is conducive to real-time understanding of the water level, water quality and meteorological conditions of each water station.

[0076] The water level prediction module includes a prediction model unit and a water level prediction unit;

[0077] The prediction model unit is used to obtain historical water conservancy data and historical meteorological data, input the historical water conservancy data and the historical meteorological data into the BP neural network for training, and use the historical water conservancy data and the historical meteorological data as input and the historical water level data as output to establish a BP neural network model;

[0078] The water level prediction unit is used to input the water conservancy data and the meteorological data of each water station into a BP neural network model, obtain the predicted water level data of each station, retrieve the predicted water level of the water station in the water level data, define the water station whose predicted water level is greater than or equal to a first expected threshold as a flood warning water station, and define the water station whose predicted water level is less than or equal to a second expected threshold as a drought warning water station;

[0079] The historical water conservancy data includes historical water level data, historical water flow data, historical evaporation data, historical water quality data and historical gate data;

[0080] The historical meteorological data includes historical temperature, historical humidity, historical air pressure, historical wind speed, historical wind direction, historical rainfall time, historical rainfall intensity and historical rainfall speed;

[0081] In this embodiment, a BP neural network model is established based on historical water conservancy data and historical meteorological data, and the BP neural network model is used to obtain predicted water level data of each water station, providing a specific and accurate scheduling basis for water conservancy scheduling based on predicted water level data and obtaining water conservancy scheduling strategies.

[0082] The water conservancy scheduling module includes a scheduling strategy unit and a simulation unit;

[0083] The scheduling strategy unit is used to obtain flood particle information based on the flood warning water station, the flood particle information including the water station type, water station location data, water station water level data and water station sluice data of other water stations except the flood warning water station, obtain the flood water station location data of the flood warning water station, obtain the flood water station scheduling distance based on the water station location data and the flood water station location data, determine the flood resistance particle value of each water station based on the flood water station scheduling distance and the water station type, obtain the water station with the largest flood resistance particle value, and define it as the flood discharge station of the flood warning water station;

[0084] Obtain drought particle information based on drought warning water stations, the drought particle information including types of water stations other than the drought warning water stations, water station location data, water level data of the water stations, and water gate data of the water stations; obtain drought water station location data of the drought warning water stations; obtain drought water station dispatching distances based on the water station location data and the drought water station location data; determine drought resistance particle values for each water station based on the drought water station dispatching distances and the water station types; obtain the water station with the largest drought resistance particle value and define it as the drought resistance water station of the drought warning water station; repeat the above steps to obtain a first water conservancy dispatching strategy;

[0085] The simulation unit is used to establish a fluid simulation model according to the water conservancy data, determine the first fluid incremental data according to the predicted water level data, and update the fluid simulation model according to the first fluid incremental data, obtain the second fluid adjustment data through the first water conservancy scheduling strategy, and input the second fluid adjustment data into the fluid simulation model to obtain the fluid simulation data, retrieve the simulated flow of the fluid simulation data, define the flow of the simulated flow greater than or equal to the first expected threshold as a first abnormal flow, define the flow of the fluid simulation model less than or equal to the second expected threshold as a second abnormal flow, define the first water conservancy scheduling strategy corresponding to the first abnormal flow or the second abnormal flow as an abnormal fluid strategy, and adjust the abnormal fluid strategy according to the first abnormal flow and the second abnormal flow to obtain the second water conservancy scheduling strategy;

[0086] Determining the drought-resistant particle value of each water station according to the drought water station scheduling distance and the water station type includes:

[0087] A scheduling distance coefficient is obtained according to the scheduling distance of the drought water station, and a scheduling water station coefficient is obtained according to the type of the water station. The scheduling water station coefficient of the flood warning water station is 1, the scheduling water station coefficient of the normal operating water station is 0, and the scheduling water station coefficient of the drought warning water station is -1. The scheduling distance coefficient and the scheduling water station coefficient are weightedly calculated to obtain the drought-resistant particle value.

[0088] Determining the flood-fighting particle value of each water station according to the flood water station dispatching distance and the water station type includes:

[0089] Obtain a dispatching distance coefficient according to the flood water station dispatching distance, and obtain a dispatching water station coefficient according to the water station type. The dispatching water station coefficient of the flood warning water station is -1, the dispatching water station coefficient of the normal operating water station is 0, and the dispatching water station coefficient of the drought warning water station is 1. Perform a weighted calculation on the dispatching distance coefficient and the dispatching water station coefficient to obtain a post-flood resistance particle value.

[0090] The types of water stations include flood warning water stations, drought warning water stations and normal operating water stations;

[0091] The fluid simulation data includes simulated water level, simulated flow rate and simulated flow velocity.

[0092] In this embodiment, a particle swarm algorithm is used to obtain a first water conservancy scheduling strategy based on flood warning water stations and drought warning water stations, diverting flood water from flood warning water stations to drought warning water stations, solving both flood and drought problems at the same time, making full use of water resources, and realizing intelligent water conservancy project scheduling.

[0093] The fluid simulation model is used to simulate the scheduling process according to the first water conservancy scheduling strategy, obtain the scheduling results, and obtain the second water conservancy scheduling strategy based on the scheduling results to improve the shortcomings of the first water conservancy scheduling strategy and ensure the feasibility of the water conservancy scheduling strategy and the safety and reliability of actual scheduling.

[0094] The hydropower generation module includes a power generation information unit and a power generation strategy unit;

[0095] The power generation information unit is used to obtain user demand information of the water station, and determine the user power supply distance according to the location of the water station and the user, and use edge computing to determine the power supply user according to the user power supply distance;

[0096] The power generation strategy unit is used to obtain the scheduling water station pair information according to the second water conservancy scheduling strategy, call the gate data of the scheduling water station pair information, and obtain the water head through the gate data, and use the water conservancy power generation calculation formula to obtain the single-gate power supply, double-gate power supply, and triple-gate power supply of the water station through the water head... The peak power consumption per unit time is obtained through the user demand information of the power supply user, and the water conservancy power generation strategy is determined according to the peak power consumption. When the peak power consumption is less than or equal to the single-gate power supply, only the single gate is opened. When the peak power consumption is greater than the single-gate power supply and less than or equal to the double-gate power supply, the double gate is opened. When the peak power consumption is greater than or equal to the double-gate power supply and less than or equal to the triple-gate power supply, the three gates are opened. Repeat the above steps to obtain the water conservancy power generation strategy, and schedule the water station gates according to the water conservancy power generation strategy.

[0097] In this embodiment, edge computing is used to obtain power supply users, and a hydropower generation strategy is determined based on user demand information and a second water conservancy scheduling strategy. Power is generated according to the hydropower generation strategy and power is supplied to power users, thereby making full use of hydropower generation and improving energy utilization.

[0098] The establishment of the BP neural network model specifically includes:

[0099] Input the historical water conservancy data and the historical meteorological data into the input layer of the BP neural network, obtain a prediction factor based on the historical water conservancy data and the historical meteorological data, randomly obtain a first weight value and a first bias value of the prediction factor, and use a first activation function to obtain a first prediction value based on the first weight value and the first bias value, use the first prediction value as the input of the hidden layer, and use a loss function to obtain a loss value based on the first prediction value and the true value, adjust the first weight value and the first bias value based on the loss value, obtain a second weight value and a second bias value, repeat the above steps n times until the loss value is less than or equal to the loss expectation threshold, then obtain a final weight value and a final bias value, and use a second activation function based on the final weight value and the final bias value to obtain predicted water level data as the output of the output layer.

[0100] In this embodiment, by establishing a BP neural network model and continuously adjusting weight values and bias values, a reliable model support is provided for obtaining predicted water level data, thereby ensuring the accuracy and reliability of the predicted values.

[0101] The water level data includes the average water level per unit time and the water level distribution per unit time;

[0102] The water flow data includes flow rate per unit time, flow velocity per unit time and runoff per unit time;

[0103] The evaporation data includes the average evaporation per unit time;

[0104] The water quality data include water temperature, turbidity, dissolved oxygen content, pH value, chemical oxygen demand, total phosphorus and total nitrogen;

[0105] The gate data includes the water level before the gate, the water level after the gate, the gate opening, the gate operating conditions and the flow rate passing through the gate.

[0106] In this embodiment, water conservancy data and meteorological data are obtained through real-time monitoring, which provides detailed and reliable data support for establishing a BP neural network surface model to predict water levels at water stations, and is conducive to real-time understanding of the water level, water quality and meteorological conditions of each water station.

[0107] The hydropower generation calculation formula is as follows:

[0108] ;

[0109] Among them, Q is the hydropower generation, q is the flow rate per unit time, H is the head value, g is the acceleration of gravity, and t is the power generation time.

[0110] In this embodiment, the hydropower generation calculation formula is used to obtain the hydropower generation amount, which provides an effective data basis for adjusting the hydropower generation strategy according to the hydropower generation amount.

[0111] Determining the power supply user according to the power supply distance of the user by using edge computing includes:

[0112] The power supply distances of the users are arranged in ascending order, and the first m users in the arrangement are selected and determined as power supply users.

[0113] In this embodiment, edge computing is used to determine power supply users based on their power supply distance, thereby improving the terminal response speed and reliability of the water conservancy project scheduling system, reducing the system's requirements for cloud server computing power and network bandwidth, and realizing the design requirements of intuitive and efficient analysis, scientific judgment, and intelligent scheduling of urban water conservancy scheduling in scenarios such as flood prevention, waterlogging control, and water circulation.

[0114] The dispatching water station pair information includes the first water station data, the second water station data and the gate data;

[0115] The user demand information includes the power consumption area number, power demand and power consumption time.

[0116] The present invention predicts the water level of a water station using a BP neural network model based on historical water conservancy data and historical meteorological data, and obtains a flood warning water station and a drought warning water station based on the prediction results. A particle swarm algorithm is used to obtain a first water conservancy scheduling strategy based on the flood warning water station and the drought warning water station, and flood water is diverted from the flood warning water station to the drought warning water station, thereby solving both flood and drought problems, making full use of water resources, and realizing intelligent water conservancy project scheduling.

[0117] The fluid simulation model is used to simulate the scheduling process according to the first water conservancy scheduling strategy, obtain the scheduling results, and obtain the second water conservancy scheduling strategy based on the scheduling results to improve the shortcomings of the first water conservancy scheduling strategy and ensure the feasibility of the water conservancy scheduling strategy and the safety and reliability of actual scheduling.

[0118] Edge computing is used to obtain power supply users, and the hydropower generation strategy is determined based on user demand information and the second water conservancy scheduling strategy. Power is generated according to the hydropower generation strategy and power is supplied to power users, which is conducive to making full use of hydropower generation and improving energy utilization.

[0119] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0120] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A water conservancy project scheduling system based on artificial intelligence, characterized in that: It includes water conservancy data monitoring module, water level prediction module, water conservancy scheduling module and water conservancy power generation module; The water conservancy data monitoring module is used to monitor and obtain water conservancy data in real time using sensors, and the meteorological monitoring platform is used to obtain meteorological data in real time; The water level prediction module is used to obtain historical water conservancy data and historical meteorological data, and establish a BP neural network model based on the historical water conservancy data and historical meteorological data, and input the water conservancy data and the meteorological data into the BP neural network model to obtain predicted water level data; The water conservancy scheduling module is used to obtain a first water conservancy scheduling strategy using a particle swarm algorithm based on the predicted water level data, establish a fluid simulation model based on the water conservancy scheduling strategy and obtain simulation results, and adjust the first water conservancy scheduling strategy based on the simulation results to obtain a second water conservancy scheduling strategy; The hydropower generation module is used to obtain user demand information, use edge computing to obtain power supply users, and determine a hydropower generation strategy based on the user demand information and the second water conservancy scheduling strategy, and perform hydropower supply for the power supply users through the hydropower generation strategy; The water conservancy scheduling module includes a scheduling strategy unit and a simulation unit; The scheduling strategy unit is used to obtain flood particle information based on the flood warning water station, the flood particle information including the water station type, water station location data, water station water level data and water station sluice data of other water stations except the flood warning water station, obtain the flood water station location data of the flood warning water station, obtain the flood water station scheduling distance based on the water station location data and the flood water station location data, determine the flood resistance particle value of each water station based on the flood water station scheduling distance and the water station type, obtain the water station with the largest flood resistance particle value, and define it as the flood discharge station of the flood warning water station; Obtain drought particle information based on drought warning water stations, the drought particle information including types of water stations other than the drought warning water stations, water station location data, water level data of the water stations, and water gate data of the water stations; obtain drought water station location data of the drought warning water stations; obtain drought water station dispatching distances based on the water station location data and the drought water station location data; determine drought resistance particle values for each water station based on the drought water station dispatching distances and the water station types; obtain the water station with the largest drought resistance particle value and define it as the drought resistance water station of the drought warning water station; repeat the above steps to obtain a first water conservancy dispatching strategy; The simulation unit is used to establish a fluid simulation model according to the water conservancy data, determine the first fluid incremental data according to the predicted water level data, and update the fluid simulation model according to the first fluid incremental data, obtain the second fluid adjustment data through the first water conservancy scheduling strategy, and input the second fluid adjustment data into the fluid simulation model to obtain the fluid simulation data, retrieve the simulated flow of the fluid simulation data, define the flow of the simulated flow greater than or equal to the first expected threshold as a first abnormal flow, define the flow of the fluid simulation model less than or equal to the second expected threshold as a second abnormal flow, define the first water conservancy scheduling strategy corresponding to the first abnormal flow or the second abnormal flow as an abnormal fluid strategy, and adjust the abnormal fluid strategy according to the first abnormal flow and the second abnormal flow to obtain the second water conservancy scheduling strategy; Determining the drought-resistant particle value of each water station according to the drought water station scheduling distance and the water station type includes: Obtain a dispatching distance coefficient based on the dispatching distance of the drought water station, and obtain a dispatching water station coefficient based on the type of the water station. The dispatching water station coefficient of the flood warning water station is 1, the dispatching water station coefficient of the normal operating water station is 0, and the dispatching water station coefficient of the drought warning water station is -1. Perform a weighted calculation on the dispatching distance coefficient and the dispatching water station coefficient to obtain a drought-resistant particle value. Determining the flood-fighting particle value of each water station according to the flood water station dispatching distance and the water station type includes: Obtain a dispatching distance coefficient according to the flood water station dispatching distance, and obtain a dispatching water station coefficient according to the water station type. The dispatching water station coefficient of the flood warning water station is -1, the dispatching water station coefficient of the normal operating water station is 0, and the dispatching water station coefficient of the drought warning water station is 1. Perform a weighted calculation on the dispatching distance coefficient and the dispatching water station coefficient to obtain a post-flood resistance particle value. The types of water stations include flood warning water stations, drought warning water stations and normal operating water stations; The fluid simulation data includes simulated water level, simulated flow rate and simulated flow velocity.

2. The artificial intelligence-based water conservancy project scheduling system according to claim 1, characterized in that: The water conservancy data monitoring module includes a water conservancy monitoring unit and a meteorological monitoring module; The water conservancy monitoring unit is used to obtain water conservancy data in real time using sensors, and the water conservancy data includes water level data, water flow data, evaporation data, water quality data and gate data; The meteorological monitoring module is used to obtain meteorological data within a future time period from the meteorological monitoring platform in real time. The meteorological data includes temperature, humidity, air pressure, wind speed, wind direction, rainfall time, rainfall intensity and rainfall speed.

3. The artificial intelligence-based water conservancy project scheduling system according to claim 1, characterized in that: The water level prediction module includes a prediction model unit and a water level prediction unit; The prediction model unit is used to obtain historical water conservancy data and historical meteorological data, input the historical water conservancy data and the historical meteorological data into the BP neural network for training, and use the historical water conservancy data and the historical meteorological data as input and the historical water level data as output to establish a BP neural network model; The water level prediction unit is used to input the water conservancy data and the meteorological data of each water station into a BP neural network model, obtain the predicted water level data of each station, retrieve the predicted water level of the water station in the water level data, define the water station whose predicted water level is greater than or equal to a first expected threshold as a flood warning water station, and define the water station whose predicted water level is less than or equal to a second expected threshold as a drought warning water station; The historical water conservancy data includes historical water level data, historical water flow data, historical evaporation data, historical water quality data and historical gate data; The historical meteorological data includes historical temperature, historical humidity, historical air pressure, historical wind speed, historical wind direction, historical rainfall time, historical rainfall intensity and historical rainfall speed.

4. The artificial intelligence-based water conservancy project scheduling system according to claim 1, characterized in that: The hydropower generation module includes a power generation information unit and a power generation strategy unit; The power generation information unit is used to obtain user demand information of the water station, and determine the user power supply distance according to the location of the water station and the user, and use edge computing to determine the power supply user according to the user power supply distance; The power generation strategy unit is used to obtain the scheduling water station pair information according to the second water conservancy scheduling strategy, call the gate data of the scheduling water station pair information, and obtain the water head through the gate data, and use the water conservancy power generation calculation formula to obtain the single-gate power supply, double-gate power supply, and triple-gate power supply of the water station through the water head... The peak power consumption per unit time is obtained through the user demand information of the power supply user, and the water conservancy power generation strategy is determined according to the peak power consumption. When the peak power consumption is less than or equal to the single-gate power supply, only the single gate is opened. When the peak power consumption is greater than the single-gate power supply and less than or equal to the double-gate power supply, the double gate is opened. When the peak power consumption is greater than or equal to the double-gate power supply and less than or equal to the triple-gate power supply, the three gates are opened. Repeat the above steps to obtain the water conservancy power generation strategy, and schedule the water station gates according to the water conservancy power generation strategy.

5. The artificial intelligence-based water conservancy project scheduling system according to claim 3, characterized in that: The establishment of the BP neural network model specifically includes: Input the historical water conservancy data and the historical meteorological data into the input layer of the BP neural network, obtain a prediction factor based on the historical water conservancy data and the historical meteorological data, randomly obtain a first weight value and a first bias value of the prediction factor, and use a first activation function to obtain a first prediction value based on the first weight value and the first bias value, use the first prediction value as the input of the hidden layer, and use a loss function to obtain a loss value based on the first prediction value and the true value, adjust the first weight value and the first bias value based on the loss value, obtain a second weight value and a second bias value, repeat the above steps n times until the loss value is less than or equal to the loss expectation threshold, then obtain a final weight value and a final bias value, and use a second activation function based on the final weight value and the final bias value to obtain predicted water level data as the output of the output layer.

6. The artificial intelligence-based water conservancy project scheduling system according to claim 2, characterized in that: The water level data includes the average water level per unit time and the water level distribution per unit time; The water flow data includes flow rate per unit time, flow velocity per unit time and runoff per unit time; The evaporation data includes the average evaporation per unit time; The water quality data include water temperature, turbidity, dissolved oxygen content, pH value, chemical oxygen demand, total phosphorus and total nitrogen; The gate data includes the water level before the gate, the water level after the gate, the gate opening, the gate operating conditions and the flow rate passing through the gate.

7. The artificial intelligence-based water conservancy project scheduling system according to claim 4, characterized in that: The hydropower generation calculation formula is as follows: ； Among them, Q is the hydropower generation, q is the flow rate per unit time, H is the head value, g is the acceleration of gravity, and t is the power generation time.

8. The artificial intelligence-based water conservancy project scheduling system according to claim 4, characterized in that: Determining the power supply user according to the power supply distance of the user by using edge computing includes: The power supply distances of the users are arranged in ascending order, and the first m users in the arrangement are selected and determined as power supply users.

9. The artificial intelligence-based water conservancy project scheduling system according to claim 4, characterized in that: The dispatching water station pair information includes the first water station data, the second water station data and the gate data; The user demand information includes the power consumption area number, power demand and power consumption time.

Images