Flexible control method and system of power grid based on wind, solar and small hydropower

Abstract

The present application discloses a method and system for flexible control of power grids based on wind, solar and small hydropower, the method comprising: constructing a head loss prediction model with historical operation data of each hydropower station in the cascade small hydropower; constructing a target hydropower prediction model with a photovoltaic power generation prediction model and a wind power generation prediction model; using each hydropower station in the cascade small hydropower as a bus topology node, using the water flow lines between each hydropower station as the data transmission line in the bus, and constructing a bus topology map with bus topology nodes and data transmission lines; constructing an optimization objective function based on the head loss prediction model, the target hydropower prediction model and the bus topology map based on minimum energy consumption and minimum path overlap rate, and outputting a hydropower station scheduling strategy. The beneficial effects of the present application: reducing the dependence of hydropower generation on a single path while reducing energy consumption, and improving the redundancy and stability of the power grid.

Description

Technical Field

[0001] The present application relates to the field of new energy dispatching technology, and in particular to a method and system for flexible power grid regulation based on wind, solar and small hydropower. Background Art

[0002] Cascade small hydropower refers to the construction of multiple small hydropower stations from upstream to downstream on rivers rich in hydropower resources according to natural conditions such as terrain and water flow, forming a cascade development pattern. The water-wind-solar complementary power station is a typical hybrid power generation system that combines photovoltaic power generation, wind power generation and hydropower generation. The existing small hydropower stations and surrounding distributed photovoltaic power stations and wind power stations are used to store and regulate electric energy through pumped storage units. When there is excess electricity, water is pumped from the lower reservoir to the upper reservoir for storage; during peak electricity consumption, water is released from the upper reservoir to the lower reservoir for power generation. This system can achieve the complementarity of hydropower, photovoltaic power and wind power at the daily, hourly and second levels. Through optimized scheduling and rapid and flexible adjustment of hydropower units, hydropower can adjust the randomly fluctuating wind power and photovoltaic power generation into smooth, stable and high-quality power sources, thereby improving the utilization rate of renewable energy and the stability of power supply.

[0003] However, the integrated water, wind and solar system involves the dispatching and operation of multiple energy sources, and an efficient dispatching system needs to be established to ensure the coordinated operation of various energy sources. In the related technology, only the leading reservoir in the cascade small hydropower is used to control and regulate the downstream water flow, which cannot achieve the reasonable allocation of each hydropower station in the cascade small hydropower, and cannot adaptively use each hydropower station in the cascade small hydropower to perform power generation in different situations. For example, if there is a large-scale reservoir-type hydropower station in the downstream hydropower station, even if the leading reservoir does not release water to the downstream, the hydropower station can still meet the power generation needs in some situations.

[0004] In addition, the development method of cascade hydropower stations is to plan a series of water conservancy hubs in the upstream and downstream of the river according to the water flow connection, which is distributed in a stepped manner. This development method allows the construction of multiple interrelated hydropower stations on a long river to form a cascade hydropower station group. However, although river tributaries are also included in the plan, in the actual water flow power generation allocation, the water flow pressure is still mostly concentrated in the mainstream. At this time, the mainstream has the problem of frequent changes in water flow, which affects the stability of the mainstream. At the same time, once the mainstream has problems, it affects the downstream power generation, and the power generation stability is poor.

[0005] The Chinese patent "A method for optimizing the dispatching diagram of cascade reservoirs based on water, wind and solar multi-energy complementarity", publication number: CN1156892418A, publication date: February 3, 2023, specifically discloses obtaining the historical runoff data of the basin, wind and solar output data, conventional power generation dispatching diagram of the reservoir and the preset power generation dispatching diagram of the reservoir, and establishing the multi-year average total power generation objective function of the water, wind and solar multi-energy complementarity system and the multi-year average power generation benefit robustness objective function of the cascade reservoir group; based on the two objective functions, after a preset set of constraints, a multi-objective optimization model of the joint dispatching diagram of the cascade reservoir group is established; based on the preset power generation dispatching diagram of the reservoir, the model is solved using a multi-objective evolutionary algorithm to obtain a non-inferior solution set of the joint dispatching diagram of the cascade reservoir group under the influence of water, wind and solar multi-energy complementarity, and determine the optimal joint dispatching diagram of the cascade reservoir group based on the water, wind and solar multi-energy complementarity dispatching demand. This technical solution still concentrates the power generation pressure on the mainstream, and the power generation stability is poor.

[0006] The Chinese patent "A method for optimizing the operation of cascade reservoirs based on the operating coefficient of cascade reservoirs", publication number: CN110942212A, publication date: March 31, 2020, specifically discloses the proposed two indicators of reservoir peak distribution coefficient and base load coefficient, determines how the remaining power of the system is distributed among the cascade hydropower stations during the study period, and how the total power generation capacity of each hydropower station is distributed in time sequence. Taking the minimum monthly maximum external power shortage rate as the optimization goal, taking into account the conditions such as the output constraints, storage capacity constraints, flow constraints, and hydraulic relationship constraints of the cascade reservoirs, an optimization model for the operating coefficient of the cascade reservoirs was established, and the solution was obtained to obtain the results of the spatiotemporal optimization allocation of cascade hydropower. Similarly, this scheme cannot achieve the adaptive adjustment of the mainstream and tributaries of cascade small hydropower, and the power generation stability is poor. Summary of the invention

[0007] In view of the problems in the prior art of hydropower station dispatching that only the tap reservoir dispatching has poor adaptability and only the mainstream dispatching has poor stability of water flow, the present application provides a flexible power grid control method and system based on wind-solar small hydropower, which uses each hydropower station in the cascade small hydropower as a bus topology node and the water flow lines between each hydropower station as the data transmission lines in the bus to construct a bus topology map, and simulates the flow of water into the flow of signals to improve the dynamic change characteristics of the topology map, so as to intuitively obtain the flow of water, thereby constructing a bus topology map with minimum energy consumption and minimum path overlap rate based on the bus topology map, the head loss prediction model and the target hydropower prediction model. The goal is to construct an optimization objective function, and obtain the optimal solution by optimizing the objective function with real-time photovoltaic power station operation data, real-time wind power station operation data and real-time hydropower station operation data, so as to execute hydropower scheduling with the hydropower station scheduling strategy with minimum energy consumption and minimum path overlap rate, reduce the dependence of hydropower generation on a single path while reducing energy consumption, avoid always using the same waterway for power generation, improve the power generation sharing capacity of tributaries and downstream large-scale water storage reservoirs, improve the redundancy and stability of the power grid, facilitate reasonable regulation according to the water flow path between reservoirs, avoid one path from bearing too much power generation pressure, and improve the rationality of hydropower scheduling.

[0008] To achieve the above technical objectives, a technical solution provided by the present application is a flexible grid control method based on wind, solar and small hydropower, including the following steps: S1: constructing a head loss prediction model based on the historical operation data of each hydropower station in the cascade small hydropower; S2: constructing a photovoltaic power generation prediction model based on the historical operation data of the photovoltaic power station, constructing a wind power generation prediction model based on the historical operation data of the wind power station, and constructing a target hydropower prediction model with the photovoltaic power generation prediction model and the wind power generation prediction model; S3: using each hydropower station in the cascade small hydropower as a bus topology node, using the water flow lines between each hydropower station as the data transmission line in the bus, and constructing a bus topology map with the bus topology nodes and the data transmission lines; S4: constructing an optimization objective function based on the minimum energy consumption and the minimum path overlap rate according to the head loss prediction model, the target hydropower prediction model and the bus topology map, so as to optimize the objective function, the real-time photovoltaic power station operation data, the real-time wind power station operation data and the real-time hydropower station operation data to output the hydropower station scheduling strategy, and execute hydropower scheduling with the hydropower station scheduling strategy.

[0009] Furthermore, S1 also includes: constructing an along-the-way head loss function based on the Darcy-Weisbach formula; constructing a local head loss function based on the local resistance coefficient; constructing a basic head loss prediction model based on the along-the-way head loss function and the local head loss function; training the basic head loss prediction model with the historical actual head, historical water flow and rated head of each hydropower station in the cascade small hydropower, and outputting the head loss prediction model.

[0010] Furthermore, S1 also includes: obtaining the measured values ​​of the along-the-way resistance coefficient and the measured values ​​of the local resistance coefficient; obtaining the training values ​​of the along-the-way resistance coefficient and the training values ​​of the local resistance coefficient in the historical time series; constructing the along-the-way resistance coefficient fluctuation curve with the measured values ​​of the along-the-way resistance coefficient and the training values ​​of the along-the-way resistance coefficient; constructing the local resistance coefficient fluctuation curve with the measured values ​​of the local resistance coefficient and the training values ​​of the local resistance coefficient; screening the special values ​​of the along-the-way resistance coefficient fluctuation curve and the local resistance coefficient fluctuation curve with the fluctuation threshold range; using the ratio of the special value to the normal value as the compensation coefficient, and retrieving the historical operation data corresponding to the special value as the compensation start-up parameter; reconstructing the head loss prediction model with the along-the-way resistance coefficient fluctuation curve including the normal value, the local resistance coefficient fluctuation curve, the compensation coefficient and the compensation start-up parameter. A fluctuation curve is constructed by measuring the measured values ​​and training values, and the fluctuation threshold range is used to take the values ​​outside the fluctuation threshold range as special values, and the values ​​within the fluctuation range threshold as normal values. The normal value is used as the normal prediction coefficient. When the power station operation data does not meet the compensation start-up parameters, the coefficient is predicted according to the normal value. When the power station operation data meets the compensation start-up parameters, the ratio of the special value to the normal value is used as the compensation coefficient. At this time, the ratio of the power station operation data to the compensation start-up parameters is calculated, and the compensation value is output based on the ratio of the power station operation data to the compensation start-up parameters and the supplementary coefficient. The coefficient predicted according to the normal value is compensated, thereby avoiding the linear compensation error caused by the sudden change of the operating state and improving the accuracy of the head loss prediction.

[0011] Furthermore, S2 includes: obtaining historical operating data of photovoltaic power stations and historical operating data of wind power stations, and performing data preprocessing; training the CNN_LSTM basic neural network model with the historical operating data of photovoltaic power stations to obtain a photovoltaic power generation prediction model; training the CNN_LSTM basic neural network model with the historical operating data of wind power stations to obtain a wind power generation prediction model; based on the target power generation power output, a complementary function of photovoltaic power generation, wind power generation and hydropower generation is constructed by fusing the complementary function, the photovoltaic power generation prediction model and the wind power generation prediction model.

[0012] Furthermore, S3 also includes: taking each hydropower station in the cascade small hydropower as a bus topology node; dividing the bus topology nodes into top-level nodes, middle-level nodes and bottom-level nodes according to the water flow regulation capacity; taking the water flow lines between each hydropower station as the data transmission lines in the bus; constructing the main flow channel, branch flow channels and control nodes according to the transmission relationship between the bus topology nodes and the data transmission lines; constructing a bus topology diagram with the top-level nodes, middle-level nodes, bottom-level nodes, main flow channels, branch flow channels and control nodes.

[0013] Furthermore, the bus topology diagram constructed by top-level nodes, middle-level nodes, bottom-level nodes, trunk channels, branch channels and control nodes also includes: constructing a point-to-point topology structure based on the trunk channels or branch channels directly connected between the top-level nodes, middle-level nodes and bottom-level nodes; constructing an arbitration ring topology structure based on the trunk channels or branch channels connected in a loop between the top-level nodes, middle-level nodes and bottom-level nodes; constructing a switching network topology structure with the top-level node as a switch; constructing a bus topology diagram with a point-to-point topology structure, an arbitration ring topology structure and a switching network topology structure. The hierarchy of cascade small hydropower is reflected by the hierarchy of the FC bus topology structure, and the various hydropower stations in the cascade small hydropower are connected in an orderly manner to construct a unified and hierarchical control network, so that hydropower stations with different regulation capabilities can be clearly distinguished, which is convenient for grasping the layout and characteristics of the entire cascade small hydropower system during subsequent scheduling.

[0014] Furthermore, the optimization objective function constructed based on minimum energy consumption and minimum path overlap rate according to the head loss prediction model, the target hydropower prediction model and the bus topology map includes: constructing the minimum energy consumption optimization objective function with the head loss and power transmission loss of the head loss prediction model; constructing the minimum path overlap rate optimization objective function with the path overlap rate of the bus topology map; constructing a dual-objective optimization function with the minimum energy consumption optimization objective function, the minimum path overlap rate optimization objective function and the target hydropower prediction value of the target hydropower prediction model.

[0015] Furthermore, the construction of the minimum energy consumption optimization objective function based on the head loss of the head loss prediction model and the power transmission loss also includes: calculating the first power transmission loss based on time series matching with the historical operation data of each hydropower station, the historical operation data of the photovoltaic power station and the historical operation data of the wind power station; obtaining the equipment power generation loss as the second power transmission loss based on the equipment operation status in the historical operation data of each hydropower station; and constructing the minimum energy consumption optimization objective function based on the head loss of the head loss prediction model and the first power transmission loss and the second power transmission loss.

[0016] Furthermore, S2 includes: constructing constraints of a photovoltaic power generation prediction model based on photovoltaic equipment; constructing constraints of a wind power generation prediction model based on wind turbine equipment; constructing reservoir capacity constraints and flow constraints of a target hydropower prediction model based on the types of each hydropower station in the cascade small hydropower; constructing grid voltage constraints and branch flow constraints based on linear flow; constructing a target hydropower prediction model with the constraints of the photovoltaic power generation prediction model, the constraints of the wind power generation prediction model, the reservoir capacity constraints, the flow constraints, the grid voltage constraints and the branch flow constraints.

[0017] Another technical solution provided by the present application is a flexible control system for a power grid based on wind, solar and small hydropower, which is used to implement the method as described above, including: a data acquisition module, used to obtain historical operating data and real-time operating data of each hydropower station, photovoltaic power station and wind power station; a data modeling module, used to construct a head loss prediction model and a target hydropower prediction model according to the historical operating data output by the data acquisition module; a topology construction module, used to construct a bus topology diagram according to the position relationship of each hydropower station in the cascade small hydropower and the water flow transmission relationship; an optimization analysis module, used to construct an optimization objective function according to the head loss prediction model, the target hydropower prediction model and the bus topology diagram, and use the optimization objective function to output the hydropower station scheduling strategy to each hydropower station in the cascade small hydropower according to the real-time operating data.

[0018] Furthermore, the data modeling module includes at least a first modeling unit for constructing a head loss prediction model, a second modeling unit for constructing a photovoltaic power generation prediction model, a third modeling unit for constructing a wind power generation prediction model, a fourth modeling unit for constructing a reservoir capacity constraint condition, a fifth modeling unit for constructing a flow constraint condition, and a sixth modeling unit for constructing a target hydropower prediction model; wherein the second modeling unit and the third modeling unit are respectively connected to the fourth modeling unit, the second modeling unit and the third modeling unit are respectively connected to the fifth modeling unit, and the fourth modeling unit and the fifth modeling unit are respectively connected to the sixth modeling unit.

[0019] The beneficial effects of the present application are as follows: a head loss prediction model is constructed based on the historical operating data of each hydropower station in the cascade small hydropower, and a photovoltaic power generation prediction model and a wind power generation prediction model are respectively constructed based on the photovoltaic power station and the wind power station that generate complementary power to the cascade small hydropower. The wind and solar power generation situation can be obtained based on the photovoltaic power generation prediction model and the wind power generation prediction model. The target hydropower prediction model is constructed based on the corresponding adjustment situation of hydropower obtained by smoothing the power fluctuations of wind and solar power generation. Each hydropower station in the cascade small hydropower is used as a bus topology node, and the water flow lines between each hydropower station are used as data transmission lines in the bus to construct a bus topology diagram. The dynamic change characteristics of the topology diagram are improved by simulating the water flow into signal flow, which facilitates dynamic regulation of the working status of each node and improves the accuracy of hydropower station regulation. At the same time, based on the bus topology diagram, head loss prediction model, and target hydropower prediction model, an optimization objective function is constructed with the minimum energy consumption and the minimum path overlap rate as the goals. The optimal solution is obtained by optimizing the objective function with real-time photovoltaic power station operation data, wind power station operation data, and hydropower station operation data. Hydropower scheduling is performed with a hydropower station scheduling strategy with minimum energy consumption and minimum path overlap rate, which reduces energy consumption while reducing the dependence of hydropower generation on a single path, improves the redundancy and stability of the power grid, and facilitates reasonable regulation according to the water flow path between reservoirs, avoiding one path from bearing too much power generation pressure, thereby improving the rationality of hydropower scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of the flexible grid control method based on wind, solar and small hydropower in this application. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the present application more clear, the present application is further described in detail below in conjunction with the drawings and examples. It should be understood that the specific implementation method described here is only an optimal embodiment of the present application, which is only used to explain the present application and does not limit the scope of protection of the present application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0022] like Figure 1 As shown, as the first embodiment of the present application, the flexible control method of the power grid based on wind, solar and small hydropower includes the following steps:

[0023] S1: Construct a head loss prediction model based on the historical operation data of each hydropower station in the cascade small hydropower;

[0024] S2: Build a photovoltaic power generation prediction model based on the historical operation data of the photovoltaic power station, build a wind power generation prediction model based on the historical operation data of the wind power station, and build a target hydropower prediction model based on the photovoltaic power generation prediction model and the wind power generation prediction model;

[0025] S3: Take each hydropower station in the cascade small hydropower as the bus topology node, take the water flow lines between each hydropower station as the data transmission line in the bus, and construct a bus topology diagram with the bus topology nodes and data transmission lines;

[0026] S4: Based on the minimum energy consumption and the minimum path overlap rate, an optimization objective function is constructed according to the head loss prediction model, the target hydropower prediction model and the bus topology diagram to optimize the objective function, the real-time photovoltaic power station operation data, the real-time wind power station operation data and the real-time hydropower station operation data to output the hydropower station scheduling strategy, and the hydropower scheduling is executed according to the hydropower station scheduling strategy.

[0027] In this embodiment, a head loss prediction model is constructed based on the historical operation data of each hydropower station in the cascade small hydropower, and a photovoltaic power generation prediction model and a wind power generation prediction model are respectively constructed based on the photovoltaic power station and the wind power station that generate complementary power with the cascade small hydropower. The wind and solar power generation situation can be obtained based on the photovoltaic power generation prediction model and the wind power generation prediction model. The target hydropower prediction model is constructed based on the corresponding adjustment of hydropower obtained by smoothing the power fluctuations of wind and solar power generation. Each hydropower station in the cascade small hydropower is used as a bus topology node, and the water flow lines between each hydropower station are used as data transmission lines in the bus to construct a bus topology diagram. The dynamic change characteristics of the topology diagram are improved by simulating the flow of water into signal flow, which facilitates dynamic regulation of the working state of each node and improves the accuracy of hydropower station regulation. At the same time, based on the bus topology diagram, head loss prediction model, and target hydropower prediction model, an optimization objective function is constructed with the minimum energy consumption and the minimum path overlap rate as the goals. The optimal solution is obtained by optimizing the objective function with real-time photovoltaic power station operation data, real-time wind power station operation data, and real-time hydropower station operation data. Hydropower scheduling is performed with a hydropower station scheduling strategy with minimum energy consumption and minimum path overlap rate, which reduces energy consumption while reducing the dependence of hydropower generation on a single path, improves the redundancy and stability of the power grid, and facilitates reasonable regulation according to the water flow path between reservoirs, avoiding one path from bearing too much power generation pressure, thereby improving the rationality of hydropower scheduling.

[0028] Head loss is the mechanical energy lost by the liquid in the process of flowing due to overcoming the water flow resistance. Head loss includes along-the-way head loss and local head loss. Therefore, step S1 also includes:

[0029] Based on the Darcy-Weisbach formula, the head loss function along the route is constructed;

[0030] Construct the local head loss function based on the local resistance coefficient;

[0031] Based on the head loss function along the way and the local head loss function, a basic prediction model for head loss is constructed;

[0032] The head loss basic prediction model is trained with the historical actual water head, historical water flow and rated water head of each hydropower station in the cascade small hydropower station, and the head loss prediction model is output.

[0033] A model is constructed based on the Darcy-Weisbach formula and the local resistance coefficient, and the model is corrected through historical operation data to improve the generalization ability and accuracy of the head loss prediction model.

[0034] Based on the Darcy-Weisbach formula, the head loss function along the route is constructed as follows:

[0035] ;

[0036] in, is the head loss along the way, is the resistance coefficient along the channel, L is the channel length, d is the hydraulic diameter of the channel, v is the average flow velocity of the section, and g is the acceleration of gravity.

[0037] The local head loss function based on the local resistance coefficient is:

[0038] ;

[0039] in, is the local resistance head loss, is the local resistance coefficient.

[0040] The basic prediction model of head loss is constructed based on the head loss function along the way and the local head loss function:

[0041] ;

[0042] in, is the local resistance coefficient of the mth local device.

[0043] The historical operation data of each hydropower station in the cascade small hydropower station includes at least: historical water flow, historical upstream and downstream water levels, and historical power generation. The historical actual water head is calculated based on the difference between the historical upstream and downstream water levels. The rated water head is the expected water head when each hydropower station is designed. At this time, the water head loss is:

[0044] ;

[0045] in, is the actual water head, is the rated water head, is the flow adjustment factor.

[0046] From this we get:

[0047] ;

[0048] Obtained through training , as well as The head loss prediction model is constructed based on the value of

[0049] In order to further improve the accuracy of the head loss prediction model, step S1 also includes:

[0050] Obtaining the measured values ​​of the resistance coefficient along the way and the measured values ​​of the local resistance coefficient;

[0051] Obtain the training values ​​of the drag coefficient along the historical time series and the training values ​​of the local drag coefficient;

[0052] The fluctuation curve of the drag coefficient along the way is constructed by using the measured value of the drag coefficient along the way and the training value of the drag coefficient along the way;

[0053] A local drag coefficient fluctuation curve is constructed based on the measured value of the local drag coefficient and the training value of the local drag coefficient;

[0054] The special values ​​of the along-the-way resistance coefficient fluctuation curve and the local resistance coefficient fluctuation curve are screened out within the fluctuation threshold range;

[0055] The ratio of the special value to the normal value is used as the compensation coefficient, and the historical operation data corresponding to the special value is retrieved as the compensation start parameter;

[0056] The head loss prediction model is reconstructed using the along-the-line resistance coefficient fluctuation curve including normal values, the local resistance coefficient fluctuation curve, the compensation coefficient and the compensation start-up parameters.

[0057] The measured values ​​of the resistance coefficient along the way and the measured values ​​of the local resistance coefficient are the channel resistance parameters and the local device resistance coefficients such as internal roughness obtained by experimental measurement methods. The fluctuation curve is constructed by the measured values ​​and the training values, and the values ​​outside the fluctuation threshold range are taken as special values ​​and the values ​​within the fluctuation range threshold are taken as normal values. The normal values ​​are used as normal prediction coefficients. When the power station operation data does not meet the compensation start parameters, the coefficients are predicted according to the normal values. When the power station operation data meets the compensation start parameters, the ratio of the special value to the normal value is used as the compensation coefficient. At this time, the ratio of the power station operation data to the compensation start parameters is calculated, and the compensation value is output by the ratio of the power station operation data to the compensation start parameters and the supplementary coefficient. The coefficients predicted according to the normal value are compensated, thereby avoiding the linear compensation error caused by the sudden change of the operating state and improving the accuracy of the head loss prediction.

[0058] The historical operation data of the photovoltaic power station includes at least: photovoltaic power generation, solar radiation intensity, temperature, humidity, wind speed and sunshine time. The historical operation data of the wind power station includes at least: wind power generation, wind direction, wind speed, temperature, humidity and air pressure. Step S2 includes:

[0059] Obtain historical operation data of photovoltaic power stations and wind power stations, and perform data preprocessing;

[0060] The photovoltaic power generation prediction model is obtained by training the CNN_LSTM basic neural network model with the historical operation data of the photovoltaic power station;

[0061] The CNN_LSTM basic neural network model is trained with the historical operation data of the wind power station to obtain the wind power generation prediction model;

[0062] Based on the complementary functions of target power generation output photovoltaic power generation, wind power generation and hydropower generation, a target hydropower prediction model is constructed by integrating the complementary function, photovoltaic power generation prediction model and wind power generation prediction model.

[0063] In this embodiment, the target value of hydropower generation is obtained through the predicted values ​​of photovoltaic power generation and wind power generation, so as to achieve compensation and stabilization of photovoltaic power generation and wind power generation through hydropower generation, and avoid power grid imbalance caused by power fluctuations when solar and wind power are connected to the grid.

[0064] The complementary functions of photovoltaic power generation, wind power generation and hydropower generation based on the target power generation output are:

[0065] ;

[0066] Where P is the target power generation, is the target power generation of the x-th hydropower unit, is the power generation power of the y-th photovoltaic device, is the power generation capacity of the z-th wind turbine, is the energy storage capacity of the x-th hydroelectric power station.

[0067] In this embodiment, when the total power generation of the photovoltaic power station and the wind power station is less than the target power generation, it means that the power generation capacity of the current photovoltaic power station and the wind power station is insufficient to meet the needs of the power system, and the power is supplemented by the hydropower station. The hydropower station adjusts its power generation output as needed, increases the power generation to supplement the shortage of photovoltaic and wind power generation, and ensures the stable operation of the power system. On the contrary, when the total power generation of the photovoltaic power station and the wind power station is greater than the target power generation, it means that the power generation capacity of the current photovoltaic power station and the wind power station exceeds the needs of the power system. In this case, in order to avoid energy waste and instability of the power grid, energy storage is performed through the hydropower station. The hydropower station uses excess electric energy to pump water into a high reservoir, converts the electric energy into water energy and stores it. In this way, when needed, water can be released to generate electricity, and the stored water energy can be converted into electric energy to meet the needs of the power system, realizing flexible scheduling and efficient utilization of energy.

[0068] The outputs of the photovoltaic power generation prediction model and the wind power generation prediction model are used as the inputs of the complementary function to obtain the target power generation parameters of the hydropower generation. It can be understood that the target power generation parameters of the hydropower generation output here are the overall power generation parameters of the cascade small hydropower. In other embodiments, the photovoltaic power generation prediction model, the wind power generation prediction model and the target hydropower prediction model can also be constructed through regression models, LSTM, support vector machines and other models.

[0069] In the cascade small hydropower system, the water flow path of each hydropower station is not limited to one main stream, and it is entirely possible that multiple tributaries merge into the main stream. Therefore, taking each hydropower station in the cascade small hydropower as a bus topology node, and the water flow lines between each hydropower station as the data transmission lines in the bus, the bus topology diagram is constructed with bus topology nodes and data transmission lines, including:

[0070] Each hydropower station in the cascade small hydropower is used as a bus topology node;

[0071] The bus topology nodes are divided into top-layer nodes, middle-layer nodes and bottom-layer nodes according to the water flow regulation capability;

[0072] The water flow lines between the various hydropower stations are used as data transmission lines in the bus;

[0073] Construct trunk channels, branch channels and control nodes according to the transmission relationship between bus topology nodes and data transmission lines;

[0074] The bus topology diagram is constructed with top-level nodes, middle-level nodes, bottom-level nodes, main flow channels, branch flow channels and control nodes.

[0075] FC bus is often used to build large-scale data transmission and communication networks. FC bus includes three main topologies: point-to-point topology, arbitration ring topology and switching topology. Based on the topological characteristics of FC bus, the flow of water is analogized to the signal transmission in FC bus, so that the dynamic change characteristics of the flow of water between cascade small hydropower stations can be displayed in the topological diagram, improving the scheduling accuracy in the subsequent hydropower station scheduling process.

[0076] In this embodiment, large reservoirs are used as top-level nodes, small and medium-sized reservoirs are used as middle-level nodes, and terminal water use points are used as bottom-level nodes. Large reservoirs and small and medium-sized reservoirs can be obtained based on the classification of reservoir levels by water conservancy projects. It can be understood that the determination of reservoir levels in water conservancy projects includes project scale, installed capacity, total storage capacity, etc. The project scale reflects the physical dimensions of reservoir construction, such as dam height, reservoir area, etc., which is an intuitive standard for measuring its water storage and regulation capabilities. The installed capacity represents the power generation potential of the reservoir hydropower station, and the total storage capacity, as the core indicator of the reservoir's water storage capacity, is directly related to its ability to regulate the spatiotemporal distribution of water flow. Therefore, the water flow regulation capacity in cascade small hydropower can be reflected based on the reservoir level.

[0077] The main flow channel is the mainstream line connecting the top-level nodes and the middle-level nodes. The branch flow channel is the branch flow channel extending from the main flow channel or the middle-level nodes, which is used to distribute water flow to different bottom-level nodes. The control nodes are gates and pump stations. By adjusting the control nodes, the water flow distribution and flow direction of different flow channels can be adjusted.

[0078] According to actual needs and water flow conditions, the working state of each control node is dynamically adjusted to achieve optimal distribution and scheduling of water flow. The hierarchical structure of the FC bus topology reflects the hierarchical nature of cascade small hydropower, and the various hydropower stations in the cascade small hydropower are connected in an orderly manner to build a unified and hierarchical control network, so that hydropower stations with different regulation capabilities can be clearly distinguished, which is convenient for grasping the layout and characteristics of the entire cascade small hydropower system during subsequent scheduling. At the same time, different hydropower stations are used as nodes at different levels. When it is necessary to add new hydropower stations or adjust the water flow path, only the corresponding nodes need to be added or replaced, which improves the modularity and scalability of hydropower scheduling, thereby improving versatility.

[0079] In this embodiment, constructing a bus topology diagram with top-level nodes, middle-level nodes, bottom-level nodes, main flow channels, branch flow channels, and control nodes also includes:

[0080] Construct a point-to-point topology structure based on the main flow channels or branch flow channels directly connected between the top-level nodes, the middle-level nodes, and the bottom-level nodes;

[0081] Construct an arbitration ring topology structure based on the main trunk or branch trunk of the loop connection between the top-level node, the middle-level node, and the bottom-level node;

[0082] Build a switched network topology with the top-level nodes as switches;

[0083] Construct bus topology diagrams with point-to-point topology, arbitration loop topology, and switched network topology.

[0084] With the top nodes, middle nodes, and bottom nodes of the main flow channel or branch flow channel that are directly connected, point-to-point connections between two nodes are constructed. At this time, the water flow can reach different nodes along these connections to realize point-to-point transmission between water flows. Multiple hydropower stations or reservoirs may be connected to each other through a series of water flow channels to form a loop, so as to construct an arbitration ring topology structure, and transmit the water flow in one direction of the loop, which is similar to the one-way transmission of data in the arbitration ring. Different flow channels are connected by control nodes to realize the exchange and distribution of water flow, and the top node is used as a switching mechanism to construct a switching network topology structure. The point-to-point topology structure, the arbitration ring topology structure, and the switching network topology structure are integrated into a bus topology diagram. At this time, the hydropower station corresponding to the large reservoir is used as the starting point of the bus, the hydropower station corresponding to the small and medium-sized reservoir is used as a relay node, and the hydropower station corresponding to the terminal water point is used as the terminal node. The main trunk and the branch flow channel form the physical path of the bus, and the control node is used as the exchange point or regulation point on the bus. It can be understood that under normal circumstances, the top reservoir in the cascade small hydropower is called the tap reservoir, which is the large reservoir defined in this embodiment. In other cases, there may be multiple large reservoirs in a cascade of small hydropower stations that can regulate the power generation capacity of the downstream cascade power stations. In this case, each large reservoir is used as a switching mechanism to construct a switching network topology to avoid ignoring the regulation capacity of the large reservoirs in the middle reaches during subsequent scheduling, thereby improving the accuracy of regulation.

[0085] The top-level nodes are used as switches to build a switched network topology including:

[0086] Construct a water flow network with top-level nodes, middle-level nodes, bottom-level nodes, main flow channels, branch flow channels, and control nodes;

[0087] The top-level node is used as the switching mechanism to connect the corresponding control nodes to build a switching network topology.

[0088] According to the regulation requirements, the top node acts as a switching mechanism to achieve water flow exchange and distribution between different flow channels by adjusting the control nodes. The switching network topology allows the water flow path and distribution to be dynamically adjusted according to actual needs, improving the flexibility of hydropower station regulation. Multiple large reservoirs act as switching mechanisms to back up and support each other, improving the reliability of the entire water resource distribution network. In other cases, the control system can also be used as a switching mechanism to control all topological nodes.

[0089] Based on the minimum energy consumption and the minimum path overlap rate, the optimization objective function is constructed according to the head loss prediction model, the target hydropower prediction model and the bus topology diagram, including:

[0090] The minimum energy consumption optimization objective function is constructed based on the head loss prediction model’s head loss and power transmission loss;

[0091] The minimum path overlap rate optimization objective function is constructed based on the path overlap rate of the bus topology graph;

[0092] A dual-objective optimization function is constructed with the minimum energy consumption optimization objective function, the minimum path overlap rate optimization objective function and the target hydropower prediction value of the target hydropower prediction model.

[0093] The output of the head loss prediction model, namely the head loss and the power transmission loss, is used as the optimization target of the minimum energy consumption optimization objective function, and the overlap rate of the path on the reproducing topology graph is used as the optimization target of the minimum path overlap optimization objective function, and a dual-objective optimization function is constructed based on this. The output prediction value of the target hydropower prediction model is used as the constraint condition of the dual-objective optimization function, that is, the total output power of the hydropower station after regulation needs to meet the output prediction value of the target hydropower prediction model under the condition of bidirectional optimization. Under the premise of meeting the needs of the power system, by considering energy consumption and path overlap rate at the same time, energy consumption and operating costs are reduced, and the stability and fault response capability of the system are improved. In the cascade small hydropower system, hydropower stations of different cascades often share the same water source, but there are multiple tributaries and main streams between two hydropower stations. By minimizing the path overlap rate, it is ensured that the water flow is fully utilized when flowing through each cascade, avoiding repeated use on the same path, and improving the overall utilization rate of water resources. At the same time, due to the certain redundancy between different cascade hydropower stations, when a cascade hydropower station fails, other cascade hydropower stations can undertake more power generation tasks, thereby improving the reliability of the system. Moreover, in cascade small hydropower systems, there is often a competitive relationship in the water flow distribution between different cascade hydropower stations. By minimizing the path overlap rate and reducing the water flow path overlap rate between different cascade hydropower stations, the reasonable distribution and scheduling of water flow can be achieved, and the power generation conflicts between different cascade hydropower stations can be reduced, thereby improving the stability of the system.

[0094] In general, the electricity from photovoltaic power stations and wind power stations is transmitted to hydropower stations through substations and output through the transmission ports of hydropower stations. Therefore, it is also necessary to consider the loss of electricity from photovoltaic power stations and wind power stations to hydropower stations and the power loss during the power generation process of hydropower stations. At this time, the minimum energy consumption optimization objective function constructed based on the head loss prediction model and the power transmission loss also includes:

[0095] Calculate the first power transmission loss based on the time series matching of the historical operation data of each hydropower station, the historical operation data of the photovoltaic power station, and the historical operation data of the wind power station;

[0096] The equipment power generation loss is obtained based on the equipment operation status in the historical operation data of each hydropower station as the second power transmission loss;

[0097] The minimum energy consumption optimization objective function is constructed based on the head loss of the head loss prediction model and the first power transmission loss and the second power transmission loss.

[0098] The minimum energy consumption optimization objective function is:

[0099] ;

[0100] Where E is the total energy consumption, is the first power transmission loss, It is the second power transmission loss.

[0101] According to the time sequence, the power receiving data of the historical operation data of each hydropower station is matched with the power generation power of the historical operation data of the photovoltaic power station and the power generation power of the historical operation data of the wind power station, so that the transmission loss of the photovoltaic power station and the wind power station in transmitting power to different hydropower stations can be obtained. The power generation loss of the hydropower station during power generation can be obtained by using the equipment operation status in the historical operation data of each hydropower station. The power generation loss includes mechanical loss, electrical loss and power conversion loss. The operating status of the turbine is obtained to calculate the mechanical loss, and the operating status of the turbine at least includes the bearing temperature and lubricating oil consumption of the turbine. The operating status of the generator is obtained to calculate the electrical loss, and the operating status of the generator at least includes the resistance, current and voltage of the generator. The power conversion loss is obtained by comparing the difference between the input water energy and the output electric energy.

[0102] The dual-objective optimization function is constructed by optimizing the minimum energy consumption objective function, optimizing the minimum path overlap rate objective function, and the target hydropower prediction value of the target hydropower prediction model:

[0103] ;

[0104] Where O is the path overlap rate.

[0105] At this time, the target hydropower prediction model output value, i.e., the target hydropower prediction value, is used to construct the constraint condition: the total hydropower power is equal to the target hydropower prediction value. It is understandable that due to the errors in the prediction and the actual operation, the fluctuation threshold range is set. At this time, the constraint condition is that the difference between the total hydropower power and the target hydropower prediction value satisfies the fluctuation threshold range. The fluctuation threshold range can be set according to the redundancy of the hydropower station's power generation or the grid-connected power suppression range. Based on the different requirements for loss and path overlap, the weight coefficient can be set accordingly to adapt to the changes in actual demand.

[0106] As a second embodiment of the present application, step S2 includes:

[0107] Constraints for building a photovoltaic power generation prediction model based on photovoltaic equipment;

[0108] Constraints for building a wind power generation prediction model based on wind turbine equipment;

[0109] Based on the types of each hydropower station in the cascade small hydropower, the reservoir capacity constraint conditions and flow constraint conditions of the target hydropower prediction model are respectively established;

[0110] Construct grid voltage constraints and branch power flow constraints based on linear power flow;

[0111] The target hydropower prediction model is constructed based on the constraints of the photovoltaic power generation prediction model, the wind power generation prediction model, the reservoir capacity constraints, the flow constraints, the grid voltage constraints and the branch flow constraints.

[0112] According to the influence of meteorological factors such as sunlight, the constraints for building a photovoltaic power generation prediction model based on photovoltaic equipment are:

[0113] ;

[0114] in, is the power generation of the y-th photovoltaic device at time t, is the upper power limit of the y-th photovoltaic device at time t, is the lower power limit of the y-th photovoltaic device at time t.

[0115] According to the influence of meteorological factors such as wind power, the constraints for building a wind power generation prediction model based on wind turbine equipment are:

[0116] ;

[0117] in, is the power generation of the zth wind turbine at time t, is the upper limit of the power of the zth wind turbine at time t, is the lower power limit of the zth wind turbine device at time t.

[0118] Hydropower units include reservoir type and runoff type. For reservoir type hydropower units, the reservoir capacity constraint conditions are constructed as follows:

[0119] ;

[0120] ;

[0121] ;

[0122] in, For the The storage capacity of a storage capacity hydropower unit at time t is: For the The upper limit of the storage capacity of a storage capacity hydropower unit at time t, For the The lower limit of the storage capacity of a storage capacity hydropower unit at time t; For the The storage capacity of a storage capacity hydropower unit at time t-1, For the The reservoir inflow of a storage capacity hydropower unit at time t is: For the The power generation flow of a reservoir-type hydropower unit at time t, For the The flood discharge of a reservoir-type hydropower unit at time t is: For the The power generation of a reservoir-type hydropower unit at time t is: For the The power generation efficiency of a reservoir-type hydropower unit.

[0123] For run-of-river hydropower units, flow constraints are established:

[0124] ;

[0125] ;

[0126] ;

[0127] in, For the The power generation of a runoff hydroelectric unit at time t is: For the The lower limit of the power generation capacity of a run-of-river hydropower unit at time t is: For the The gear value of a runoff hydropower unit at time t, For the The gear power difference of each radial flow hydropower unit is For the The power generation capacity of a run-of-river hydroelectric unit, For the The reservoir inflow of a storage capacity hydropower unit at time t is: For the The flood discharge of a reservoir-type hydropower unit at time t.

[0128] Based on the linear power flow model, the voltage amplitude of each node in the distribution network, the current of each branch and the active power injected into the grid connection point are:

[0129] ;

[0130] ;

[0131] ;

[0132] in, is the vector formed by the amplitude of each node and each phase voltage at time t, is the vector formed by the power of branch s and branch d at time t, is the injected active power at the distribution network connection point at time t, C, D, G are constant matrices related to the power grid network topology, , is a constant vector related to the power grid network topology at time t, is a constant related to the power grid topology at time t, is the vector formed by the active power injected by each node at time t. Based on the linear power flow model, the system constant parameters can be obtained.

[0133] The grid voltage constraint and branch power flow constraint based on linear power flow are:

[0134] ;

[0135] ;

[0136] in, is the vector formed by the minimum voltage of each phase and each node, is the vector formed by the maximum voltage of each phase and each node, It is a vector formed by the maximum value of active power flow of each branch.

[0137] The power aggregate of wind, solar and small hydropower is equivalent to an equivalent generator model:

[0138] ;

[0139] in, is the power generated by the equivalent generator model at time t, is the upper bound of the active power of the equivalent generator model at time t, is the lower bound of the active power of the equivalent generator model at time t.

[0140] Then, the target hydropower prediction model is constructed based on the constraints of the photovoltaic power generation prediction model, the wind power generation prediction model, the reservoir capacity constraints, the flow constraints, the grid voltage constraints and the branch flow constraints:

[0141] ;

[0142] ;

[0143] ;

[0144] ;

[0145] ;

[0146] ;

[0147] ;

[0148] ;

[0149] ;

[0150] ;

[0151] ;

[0152] ;

[0153] The solution is solved by a constraint column generation algorithm, and the prediction value of the target hydropower prediction model is obtained with the real-time photovoltaic power station operation data, wind power station operation data and hydropower station operation data. At this time, the real-time photovoltaic power station operation data at least includes real-time meteorological data and real-time photovoltaic power station equipment operation parameters, the real-time wind power station operation data at least includes real-time meteorological data and real-time wind power station equipment operation parameters, and the real-time hydropower station operation data at least includes real-time water flow data and real-time hydropower station equipment operation parameters.

[0154] As a third embodiment of the present application, a flexible control system for a power grid based on wind, solar and small hydropower is connected to each hydropower station in a cascade of small hydropower stations, including:

[0155] Data acquisition module, used to obtain historical and real-time operation data of each hydropower station, photovoltaic power station and wind power station;

[0156] A data modeling module is used to construct a head loss prediction model and a target hydropower prediction model according to the historical operation data output by the data acquisition module;

[0157] A topology building module is used to build a bus topology diagram based on the position relationship of each cascade small hydropower station and the water flow transmission relationship;

[0158] The optimization analysis module is used to construct an optimization objective function based on the head loss prediction model, the target hydropower prediction model and the bus topology diagram, and use the optimization objective function to output the hydropower station scheduling strategy to each hydropower station in the cascade small hydropower according to the real-time operation data.

[0159] In this embodiment, the data acquisition module is connected to the data modeling module, the topology construction module and the optimization analysis module, the data modeling module is connected to the optimization analysis module, and the topology construction module is connected to the optimization analysis module. By constructing a head loss prediction model, a target hydropower prediction model and a bus topology diagram, the water flow loss relationship, the wind-solar-hydropower generation complementary relationship and the water flow transmission relationship are obtained, and based on the above relationship, a two-way optimization of minimum energy consumption and minimum path overlap rate is performed to obtain a hydropower station scheduling strategy that satisfies energy consumption and avoids a main path from bearing too much water flow change pressure. While reducing the complementary loss of power generation, the coordination between cascaded small hydropower stations is improved, the power generation pressure between cascaded small hydropower stations is balanced, and the rationality of hydropower generation is improved.

[0160] The data modeling module at least includes a first modeling unit for constructing a head loss prediction model, a second modeling unit for constructing a photovoltaic power generation prediction model, a third modeling unit for constructing a wind power generation prediction model, and a fourth modeling unit for constructing a target hydropower prediction model. The second modeling unit and the third modeling unit are respectively connected to the fourth modeling unit.

[0161] As the fourth embodiment of the present application, the data modeling module at least includes a first modeling unit for constructing a head loss prediction model, a second modeling unit for constructing a photovoltaic power generation prediction model, a third modeling unit for constructing a wind power generation prediction model, a fourth modeling unit for constructing a reservoir capacity constraint condition, a fifth modeling unit for constructing a flow constraint condition, and a sixth modeling unit for constructing a target hydropower prediction model. Among them, the second modeling unit and the third modeling unit are respectively connected to the fourth modeling unit, the second modeling unit and the third modeling unit are respectively connected to the fifth modeling unit, and the fourth modeling unit and the fifth modeling unit are respectively connected to the sixth modeling unit.

[0162] As the fifth embodiment of the present application, the electronic device includes a processor, a memory and a battery module; the memory is used to store programs; the battery module is used to power the memory; the processor is used to execute the program and implement the above-mentioned flexible power grid control method based on wind, solar and small hydropower when executing the program.

[0163] As the sixth embodiment of the present application, a computer-readable storage medium is used to store a computer program or instruction. When the computer program or instruction is executed by a processing device, the above-mentioned flexible control method of the power grid based on wind, solar and small hydropower is implemented. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state hard disk), etc.

[0164] The specific implementation method described above is a preferred implementation method of the flexible grid control method and system based on wind, solar and small hydropower of this application, and is not intended to limit the specific implementation scope of this application. The scope of this application includes but is not limited to this specific implementation method. All equivalent changes made in accordance with the shape and structure of this application are within the protection scope of this application.

Claims

1. A flexible control method for power grid based on wind, solar and small hydropower, characterized by: The steps include: S1: Construct a head loss prediction model based on the historical operation data of each hydropower station in the cascade small hydropower; S2: Build a photovoltaic power generation prediction model based on the historical operation data of the photovoltaic power station, build a wind power generation prediction model based on the historical operation data of the wind power station, and build a target hydropower prediction model based on the photovoltaic power generation prediction model and the wind power generation prediction model; S3: Take each hydropower station in the cascade small hydropower as the bus topology node, take the water flow lines between each hydropower station as the data transmission line in the bus, and construct a bus topology diagram with the bus topology nodes and data transmission lines; S4: Based on the minimum energy consumption and the minimum path overlap rate, an optimization objective function is constructed according to the head loss prediction model, the target hydropower prediction model and the bus topology diagram, so as to optimize the objective function, the real-time photovoltaic power station operation data, the real-time wind power station operation data and the real-time hydropower station operation data to output the hydropower station scheduling strategy, and execute the hydropower scheduling according to the hydropower station scheduling strategy; The S2 includes: Obtain historical operation data of photovoltaic power stations and wind power stations, and perform data preprocessing; The photovoltaic power generation prediction model is obtained by training the CNN_LSTM basic neural network model with the historical operation data of the photovoltaic power station; The CNN_LSTM basic neural network model is trained with the historical operation data of the wind power station to obtain the wind power generation prediction model; Based on the complementary functions of photovoltaic power generation, wind power generation and hydropower generation outputted by the target power generation, the target hydropower prediction model is constructed by integrating the complementary functions, the photovoltaic power generation prediction model and the wind power generation prediction model; The optimization objective function constructed based on the minimum energy consumption and the minimum path overlap rate according to the head loss prediction model, the target hydropower prediction model and the bus topology diagram includes: The minimum energy consumption optimization objective function is constructed based on the head loss prediction model’s head loss and power transmission loss; The minimum path overlap rate optimization objective function is constructed based on the path overlap rate of the bus topology graph; A dual-objective optimization function is constructed with the minimum energy consumption optimization objective function, the minimum path overlap rate optimization objective function and the target hydropower prediction value of the target hydropower prediction model.

2. The method for flexible control of power grid based on wind, solar and small hydropower as claimed in claim 1, characterized in that: The S1 further comprises: Based on the Darcy-Weisbach formula, the head loss function along the route is constructed; Construct the local head loss function based on the local resistance coefficient; Based on the head loss function along the way and the local head loss function, a basic prediction model for head loss is constructed; The head loss basic prediction model is trained with the historical actual water head, historical water flow and rated water head of each hydropower station in the cascade small hydropower station, and the head loss prediction model is output.

3. The method for flexible control of power grid based on wind, solar and small hydropower as claimed in claim 2, characterized in that: The S1 further comprises: Obtaining the measured values ​​of the resistance coefficient along the way and the measured values ​​of the local resistance coefficient; Obtain the training values ​​of the drag coefficient along the historical time series and the training values ​​of the local drag coefficient; The fluctuation curve of the drag coefficient along the way is constructed by using the measured value of the drag coefficient along the way and the training value of the drag coefficient along the way; A local drag coefficient fluctuation curve is constructed based on the measured value of the local drag coefficient and the training value of the local drag coefficient; The special values ​​of the along-the-way resistance coefficient fluctuation curve and the local resistance coefficient fluctuation curve are screened out within the fluctuation threshold range; The ratio of the special value to the normal value is used as the compensation coefficient, and the historical operation data corresponding to the special value is retrieved as the compensation start parameter; The head loss prediction model is reconstructed using the along-the-line resistance coefficient fluctuation curve including normal values, the local resistance coefficient fluctuation curve, the compensation coefficient and the compensation start-up parameters.

4. The method for flexible control of power grid based on wind, solar and small hydropower as claimed in claim 1, characterized in that: The S3 further includes: Each hydropower station in the cascade small hydropower is used as a bus topology node; The bus topology nodes are divided into top-layer nodes, middle-layer nodes and bottom-layer nodes according to the water flow regulation capability; The water flow lines between the various hydropower stations are used as data transmission lines in the bus; Construct trunk channels, branch channels and control nodes according to the transmission relationship between bus topology nodes and data transmission lines; The bus topology diagram is constructed with top-level nodes, middle-level nodes, bottom-level nodes, main flow channels, branch flow channels and control nodes.

5. The method for flexible control of power grid based on wind, solar and small hydropower as claimed in claim 4, characterized in that: The bus topology diagram constructed by using top-level nodes, middle-level nodes, bottom-level nodes, main flow channels, branch flow channels and control nodes also includes: Construct a point-to-point topology structure based on the main flow channels or branch flow channels directly connected between the top-level nodes, the middle-level nodes, and the bottom-level nodes; Construct an arbitration ring topology structure based on the main trunk or branch trunk of the loop connection between the top-level node, the middle-level node, and the bottom-level node; Build a switched network topology with the top-level nodes as switches; Construct bus topology diagrams with point-to-point topology, arbitration loop topology, and switched network topology.

6. The method for flexible control of power grid based on wind, solar and small hydropower as claimed in claim 1, characterized in that: The minimum energy consumption optimization objective function constructed by using the head loss prediction model and the power transmission loss also includes: Calculate the first power transmission loss based on the time series matching of the historical operation data of each hydropower station, the historical operation data of the photovoltaic power station, and the historical operation data of the wind power station; The equipment power generation loss is obtained based on the equipment operation status in the historical operation data of each hydropower station as the second power transmission loss; The minimum energy consumption optimization objective function is constructed based on the head loss of the head loss prediction model and the first power transmission loss and the second power transmission loss.

7. The method for flexible control of power grid based on wind, solar and small hydropower as claimed in claim 1, characterized in that: The S2 includes: Constraints for building a photovoltaic power generation prediction model based on photovoltaic equipment; Constraints for building a wind power generation prediction model based on wind turbine equipment; Based on the types of each hydropower station in the cascade small hydropower, the reservoir capacity constraint conditions and flow constraint conditions of the target hydropower prediction model are respectively established; Construct grid voltage constraints and branch power flow constraints based on linear power flow; The target hydropower prediction model is constructed based on the constraints of the photovoltaic power generation prediction model, the wind power generation prediction model, the reservoir capacity constraints, the flow constraints, the grid voltage constraints and the branch flow constraints.

8. A flexible control system for power grid based on wind, solar and small hydropower, used to implement the method according to any one of claims 1 to 7, characterized in that: include: Data acquisition module, used to obtain historical and real-time operation data of each hydropower station, photovoltaic power station and wind power station; A data modeling module is used to construct a head loss prediction model and a target hydropower prediction model according to the historical operation data output by the data acquisition module; A topology building module is used to build a bus topology diagram based on the position relationship of each cascade small hydropower station and the water flow transmission relationship; The optimization analysis module is used to construct an optimization objective function based on the head loss prediction model, the target hydropower prediction model and the bus topology diagram, and use the optimization objective function to output the hydropower station scheduling strategy to each hydropower station in the cascade small hydropower according to the real-time operation data.

9. The flexible control system for wind, solar and small hydropower based power grid according to claim 8, characterized in that: The data modeling module at least includes a first modeling unit for constructing a head loss prediction model, a second modeling unit for constructing a photovoltaic power generation prediction model, a third modeling unit for constructing a wind power generation prediction model, a fourth modeling unit for constructing a reservoir capacity constraint condition, a fifth modeling unit for constructing a flow constraint condition, and a sixth modeling unit for constructing a target hydropower prediction model; Among them, the second modeling unit and the third modeling unit are connected to the fourth modeling unit respectively, the second modeling unit and the third modeling unit are connected to the fifth modeling unit respectively, and the fourth modeling unit and the fifth modeling unit are connected to the sixth modeling unit respectively.

Images