Cross-basin hydropower station group short-term optimization scheduling method considering virtual scheduling partition

Abstract

The invention discloses an inter-basin hydropower station group short-term optimization scheduling method considering virtual scheduling partitions. According to the method provided by the invention, by introducing a virtual scheduling partitioning mechanism, a complex cross-basin hydropower station group system is deconstructed into a plurality of subunits which operate cooperatively, firstly, reasonable division of the hydropower station group is realized according to a hydraulic power and electric power coupling relationship so as to remarkably reduce the dimensionality of an optimization problem, and the problem of curse of dimensionality is effectively avoided. And the real-time supply and demand characteristics and the adjustment potential of different areas are accurately identified by dynamically judging the input / output attributes of each subarea. And finally, collaborative optimization is carried out by taking the partitions as basic scheduling units, and the overall operation benefit of the system is maximized on the premise of ensuring that various complex constraints are met, so that the synchronous improvement of the scheduling calculation efficiency and the cross-basin hydropower collaborative complementary benefit is realized on the whole.

Description

Technical Field

[0001] This invention relates to the technical field of hydropower station group scheduling optimization, and specifically to a short-term optimization scheduling method for cross-basin hydropower station groups that considers virtual scheduling partitions. Background Technology

[0002] In related technologies, with the increasing frequency of extreme weather events and the rising proportion of clean energy sources such as wind and solar power in recent years, the joint dispatch of inter-basin hydropower station groups has become a key means to enhance the grid's capacity to absorb new energy and ensure the safe and economical operation of the grid. However, the rapid growth in the number and scale of hydropower stations and the complex hydraulic and electrical coupling relationships between river basins have brought enormous challenges to the short-term optimal dispatch of inter-basin hydropower station groups.

[0003] Existing research on short-term hydropower station dispatch mostly focuses on the optimization of cascade hydropower stations within a single river basin. While such methods help maximize the benefits for power generation companies or meet local grid peak-shaving targets, their overall dispatching capacity and inter-regional coordination flexibility are significantly insufficient when facing the power supply guarantee pressure of provincial or even larger-scale power grids. Therefore, there is an urgent need to conduct research on the coordinated dispatching of inter-basin hydropower station groups to improve the grid's ability to ensure power supply under various extreme conditions. Reasonable division of hydropower station groups can effectively reduce dispatching dimensionality and improve optimization efficiency. However, existing research on power station group partitioning mostly focuses on wind and solar power station groups or microgrid systems, and their partitioning indicators often emphasize the electrical connections between power stations, such as active / reactive power balance and power coupling. They rarely consider the impact of the unique and complex hydraulic connections between hydropower stations on the partitioning results, making existing methods difficult to directly apply to the virtual partitioning requirements of hydropower station groups. Furthermore, at the optimization method level, existing dispatching models mostly adopt unified centralized optimization or simple decomposition and coordination strategies, failing to fully consider the spatiotemporal heterogeneity and functional differentiation characteristics existing in the operation of inter-basin systems. Specifically, the system naturally forms a "power output zone" with a power surplus and a "power input zone" with a power deficit during different operating periods. Existing "one-size-fits-all" optimization models cannot accurately identify and respond to such dynamic changes in functional attributes, making it difficult for scheduling strategies to achieve precise matching and optimized allocation of power on a spatiotemporal scale. This limits the scale benefits and complementary potential that a cross-basin hydropower station group should possess.

[0004] In summary, among the relevant technologies, the short-term optimal scheduling of cross-basin hydropower station groups faces technical challenges such as high computational complexity and insufficient cross-regional coordination capabilities due to the high system dimensionality and complex coupling relationships. Summary of the Invention

[0005] The technical problem this invention aims to solve is that, in related technologies, the short-term optimal scheduling of inter-basin hydropower station groups faces challenges such as high computational complexity and insufficient cross-regional coordination capabilities due to the high system dimensionality and complex coupling relationships. The purpose is to provide a short-term optimal scheduling method for inter-basin hydropower station groups that considers virtual scheduling partitions. This solves the technical problems of high computational complexity and insufficient cross-regional coordination capabilities.

[0006] This invention is achieved through the following technical solution:

[0007] In a first aspect, the present invention provides a short-term optimal scheduling method for cross-basin hydropower station groups considering virtual scheduling partitions, comprising the following steps:

[0008] Based on the strength of hydraulic and electrical connections between hydropower stations in the cross-basin hydropower station group, the hydropower station group is divided into multiple virtual scheduling zones.

[0009] For each virtual dispatching partition, based on the power supply and demand balance conditions of the partition and the end-of-day water level constraints of the hydropower station, the virtual dispatching partition is identified as an input partition or an output partition; wherein, the input partition indicates a power shortage that requires power reception, and the output partition indicates a power surplus that can be transmitted to other regions;

[0010] Using the virtual scheduling partition as the scheduling unit, a preset optimization algorithm is used to solve the preset short-term optimization scheduling model to obtain the short-term scheduling plan of the cross-basin hydropower station group; wherein, the short-term optimization scheduling model is configured to take the maximization of the total system benefit as the objective function, and includes at least the inter-regional power balance constraints and the hydropower station operation constraints within the partition.

[0011] Furthermore, the step of dividing the hydropower station group into multiple virtual scheduling zones based on the hydraulic and electrical connection strengths between the hydropower stations in the inter-basin hydropower station group includes:

[0012] For any two hydropower stations, calculate the power connection strength index and the hydraulic connection strength index between them respectively;

[0013] The electrical connection strength index and the hydraulic connection strength index are normalized and then weighted and fused to obtain the comprehensive connection strength.

[0014] A complex network of hydropower stations is constructed, with each hydropower station as a node and the comprehensive connection strength as the edge weight.

[0015] Based on the principle of maximizing modularity, the complex network of the hydropower station is divided into communities, and the resulting communities are used as the virtual scheduling partitions.

[0016] Furthermore, the step of calculating the power linkage strength index and hydraulic linkage strength index between any two hydropower stations includes:

[0017] Based on the sensitivity analysis of the power grid, the electrical distance reflecting the electrical coupling relationship between the nodes of each hydropower station is determined as the power connection strength index; wherein, the electrical distance is negatively correlated with the power connection strength between two hydropower stations.

[0018] Based on the topological relationship between hydropower stations and the regulation characteristics of reservoirs, a comprehensive factor reflecting their hydraulic coupling relationship is determined as an index of hydraulic connection strength; wherein, the comprehensive factor includes at least a flow lag factor reflecting the water flow propagation time and a reservoir capacity regulation factor reflecting the reservoir regulation capacity.

[0019] Furthermore, the step of dividing the complex network of the hydropower station into communities based on the principle of maximizing modularity, and using the resulting communities as the virtual scheduling partitions, includes:

[0020] A heuristic algorithm based on modularity gain is used to iteratively detect communities in the complex network of the hydropower station until the network modularity no longer increases, so as to obtain the virtual scheduling partition.

[0021] Furthermore, the step of determining whether a virtual scheduling partition is an input partition or an output partition based on the power supply and demand balance conditions and the end-of-day water level constraints of the hydropower station for each virtual scheduling partition includes:

[0022] If the total real-time output of a virtual dispatching zone during the dispatching period is greater than or equal to its load demand, and the water level at the end of the dispatching period of each hydropower station in the zone is not lower than its preset end-of-day water level boundary condition, then the zone is determined to be an output-type zone.

[0023] If the total real-time output of a virtual dispatching zone during the dispatching period is less than its load demand, or if the water level at the end of the dispatching period of any hydropower station in the zone is lower than its preset end-of-day water level boundary condition in order to meet the load demand, then the zone is determined to be an input-type zone.

[0024] Furthermore, the objective function of the preset short-term optimization scheduling model is configured as follows:

[0025] Maximize the net difference between the total electricity sales revenue of all output-type zones and the total electricity purchase cost of all input-type zones during the dispatch period;

[0026] The pre-defined short-term optimization scheduling model's interval power balance constraint is configured such that the sum of the net output of all output-type zones during any scheduling period is equal to the sum of the net load demand of all input-type zones during the same period.

[0027] The operational constraints of the hydropower stations within the specified zone are configured as follows: water balance constraints, power generation flow constraints, discharge flow constraints, reservoir capacity constraints, hydraulic connection constraints, power output constraints, water level fluctuation constraints, and initial and final water level constraints.

[0028] Further, the step of using the virtual scheduling partition as the scheduling unit and employing a preset optimization algorithm to solve the preset short-term optimization scheduling model to obtain the short-term scheduling plan for the cross-basin hydropower station group includes:

[0029] The short-term optimal scheduling model is solved using a stepwise optimization algorithm. The multi-stage decision problem for the entire scheduling period is decomposed into a series of sequentially executed two-stage optimization sub-problems. Each two-stage sub-problem is optimized and solved, and the short-term scheduling plan is obtained through iterative calculation.

[0030] Secondly, the present invention provides a short-term optimized scheduling device for cross-basin hydropower station groups considering virtual scheduling partitions, comprising:

[0031] The partitioning module is used to divide the hydropower station group into multiple virtual scheduling zones based on the hydraulic and electrical connection strengths between the hydropower stations in the cross-basin hydropower station group.

[0032] The discrimination module is used to classify each virtual scheduling partition as an input partition or an output partition based on the power supply and demand balance conditions and the end-of-day water level constraints of the hydropower station. The input partition indicates a power shortage that requires power supply, while the output partition indicates a power surplus that can be transmitted to other regions.

[0033] The solution module is used to solve the preset short-term optimization scheduling model using the virtual scheduling partition as the scheduling unit and a preset optimization algorithm to obtain the short-term scheduling plan of the cross-basin hydropower station group; wherein the short-term optimization scheduling model is configured to take the maximization of the total system benefit as the objective function, and includes at least the inter-regional power balance constraints and the hydropower station operation constraints within the partition.

[0034] Thirdly, the present invention provides an electronic device, comprising: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, the instructions being executed by the one or more processors to cause the one or more processors to implement the method described above.

[0035] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0036] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0037] The method provided by this invention deconstructs a complex inter-basin hydropower station system into multiple collaboratively operating sub-units by introducing a virtual scheduling partitioning mechanism. First, it rationally divides the power station group based on the coupling relationship between hydraulics and electricity to significantly reduce the dimensionality of the optimization problem and effectively avoid the curse of dimensionality. Then, by dynamically determining the input / output attributes of each partition, it accurately identifies the real-time supply and demand characteristics and regulation potential of different regions. Finally, it performs collaborative optimization using partitions as the basic scheduling unit, maximizing the overall system operational efficiency while ensuring that various complex constraints are met. This achieves a simultaneous improvement in both scheduling computation efficiency and the complementary benefits of inter-basin hydropower collaboration. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0039] Figure 1 A flowchart illustrating a short-term optimized scheduling method for a cross-basin hydropower station group considering virtual scheduling partitions, provided in the embodiments of this specification;

[0040] Figure 2 This is a flowchart of the virtual scheduling partition solution for hydropower station groups provided in the embodiments of this specification;

[0041] Figure 3 This is a schematic diagram of a virtual partitioning of a cross-basin hydropower station group under complex hydraulic-electric coupling constraints, as provided in the embodiments of this specification.

[0042] Figure 4 This is a comparison chart showing the efficiency of unpartitioned solutions and solutions considering virtual partitions in the embodiments provided in this specification. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0044] In related technologies, with the continuous increase in the proportion of clean energy and the continuous expansion of the power grid, the joint optimization and dispatch of inter-basin hydropower station groups has become a key means to ensure the safe, stable, and economical operation of the power system. Such systems typically include: a central dispatch center, numerous hydropower stations distributed across different river basins, a transmission network connecting each station to the load center, and corresponding data acquisition and monitoring systems. The central dispatch center can obtain real-time information on the entire network's water conditions, power load, and unit status through the communication network, and generate dispatch plans based on optimization models, which are then distributed to each hydropower station for execution.

[0045] In related technologies, for short-term (daily or weekly) optimal scheduling of inter-basin hydropower station groups, centralized optimization or rule-based decomposition scheduling strategies are often employed. Centralized optimization treats all hydropower stations within a basin as a whole, constructing a high-dimensional, nonlinear optimization model that includes the operational constraints of all stations for unified solution. Theoretically, this method can find the global optimum. However, when there are numerous hydropower stations and complex inter-basin hydraulic and electrical coupling relationships, the scale of decision variables and constraints expands dramatically, leading to an exponential increase in computational complexity and the curse of dimensionality. This results in an excessively long solution process, making it difficult to meet the real-time decision-making requirements of short-term scheduling.

[0046] On the other hand, rule-based decomposition methods (e.g., by river basin or administrative division), while reducing the scale of the problem, often overlook the actual, dynamically changing coupling strength between power plants. In particular, they fail to effectively identify and respond to dynamic power surplus and deficit areas naturally formed within the power grid due to load changes, fluctuations in renewable energy output, and differences in reservoir regulation capacity. This makes it difficult for dispatch strategies to fully utilize the spatiotemporal complementarity across river basins, limiting the full realization of the scale advantages and synergistic benefits of hydropower plant clusters, resulting in losses in overall system operating efficiency and economy.

[0047] The root cause of the aforementioned technical problems lies in the lack of a systematic framework that can effectively balance computational efficiency and optimization accuracy. On the one hand, centralized optimization, while comprehensive, is constrained by computational bottlenecks; on the other hand, simple decomposition methods, while fast, lack sufficient collaborative capabilities. The deeper deficiency lies in the failure to rationally structure and decompose the complex network-wide scheduling problem, that is, the failure to aggregate physically or electrically connected power plants into collaborative scheduling units, and the failure to assign clear dynamic roles and interaction rules to these units based on the real-time operating status of the system.

[0048] In view of this, the concept of this invention is to propose a hierarchical and progressive optimization scheduling method by introducing the concept of virtual scheduling partitions. This method first automatically divides a large hydropower station group into multiple internally tightly coupled virtual scheduling partitions based on the actual hydraulic and electrical connection strength between hydropower stations, significantly reducing the problem's dimensionality. Then, for each partition, its role in the power grid—whether a power output area or a power input area—is dynamically determined based on its real-time power balance and the conditions for sustainable hydropower operation. Finally, using these virtual partitions as basic scheduling units, a collaborative optimization model aimed at maximizing the overall system benefit is established and efficiently solved. This concept effectively solves the technical problems of high computational complexity and insufficient cross-regional collaborative capabilities caused by the high system dimensionality and complex coupling relationships.

[0049] This embodiment provides a short-term optimized scheduling method for cross-basin hydropower station groups that considers virtual scheduling partitions. The method is executed by the controller of the hydropower station group scheduling system. The controller of the hydropower station group scheduling system can be a high-performance server, an industrial control computer designed for an industrial control environment, or a master station control unit integrated in a distributed control system, etc.

[0050] like Figure 1 and Figure 2 As shown, the method may include the following steps:

[0051] Step S12: Based on the hydraulic and electrical connection strengths between the hydropower stations in the cross-basin hydropower station group, the hydropower station group is divided into multiple virtual scheduling zones.

[0052] In this implementation, the executing entity can reorganize the original set of hydropower stations into several internally tightly coupled but externally relatively independent subsets, i.e., virtual scheduling partitions, based on predefined correlation metrics.

[0053] In this embodiment, the power connection strength can be characterized by calculating the electrical distance between hydropower station nodes and the power grid. For example, based on the active / reactive voltage sensitivity matrix obtained from power flow calculations, an index reflecting the degree of voltage mutual influence between any two nodes can be derived. The reciprocal or complementary value of this index can be used as a specific form of power connection strength; the closer the electrical distance, the higher the power connection strength.

[0054] The hydraulic connection strength can include a flow lag factor and a reservoir capacity regulation factor. Specifically, the flow lag factor can be used to characterize the time delay in the propagation of the discharge from the upstream power station to the downstream power station. The shorter the delay, the larger the value of this factor, indicating a stronger hydraulic connection. The reservoir capacity regulation factor can be used to characterize the impact of the upstream reservoir's regulation capacity on the operation of the downstream power station. The larger the effective regulation capacity of the upstream reservoir, the larger the value of this factor can be, indicating a greater potential for coordinated operation through reservoir scheduling. The final hydraulic connection strength can be a weighted combination of the above factors.

[0055] In this implementation, after determining the electrical and hydraulic connection strength data among all power stations, the executing entity can normalize the data to eliminate the influence of dimensions and selectively perform weighted fusion to form a comprehensive connection strength index. Then, using this comprehensive connection strength as edge weights, a weighted undirected network graph with hydropower stations as nodes can be constructed. Finally, a community discovery algorithm (e.g., a heuristic iterative algorithm based on modularity gain) can be applied to partition the network. The core process of this algorithm is to continuously attempt to merge nodes or communities to maximize or nearly maximize the modularity index of the entire network, thereby obtaining the final virtual scheduling partitioning scheme.

[0056] Step S14: For each virtual scheduling partition, based on the power supply and demand balance conditions of the partition and the end-of-day water level constraints of the hydropower station, the virtual scheduling partition is identified as an input partition or an output partition; wherein, the input partition indicates that the power shortage requires power supply, and the output partition indicates that the power surplus can be transmitted.

[0057] In this embodiment, the power supply and demand balance condition can be expressed as the matching relationship between power generation output and load demand within a zone. For example, it can specifically be the comparison between the total output of all controllable and uncontrollable power sources (hydropower stations, wind farms, photovoltaic power stations) within the zone and the net load demand (after deducting the basic load borne by thermal power, etc.) assigned to the zone by the power grid.

[0058] In this embodiment, the end-of-day water level constraint of the hydropower station can be expressed as the water level limit set at the end of the scheduling period to ensure the comprehensive utilization benefits of the reservoir (e.g., flood control, navigation, ecology, etc.), which can specifically be the highest and lowest allowed operating water level boundaries of each reservoir.

[0059] In one possible and specific implementation scheme, for each virtual dispatching zone, based on the zone's power supply and demand balance conditions and the hydropower station's end-of-day water level constraints, the virtual dispatching zone can be classified as either an input or output zone as follows: If a zone's total real-time output is not less than its net load demand in all time periods within the dispatching cycle, and the actual water levels of all its hydropower stations reach or exceed their respective end-of-day water level boundary conditions at the end of the dispatching period, then the zone can be classified as an output zone. That is, the zone has surplus power to supply electricity after meeting its own needs. Conversely, if a zone's total real-time output is insufficient to meet its net load demand in any dispatching period, or even if the load in each time period is met by adjusting hydropower output, but at least one hydropower station in the zone has a water level below its end-of-day water level boundary condition at the end of the dispatching period, then the zone is classified as an input zone. This indicates that the zone needs to receive electricity from external sources to reliably complete the dispatching task or avoid disrupting the sustainable operation of the reservoir.

[0060] Step S16: Using the virtual scheduling partition as the scheduling unit, the preset short-term optimization scheduling model is solved by a preset optimization algorithm to obtain the short-term scheduling plan of the cross-basin hydropower station group; wherein, the short-term optimization scheduling model is configured to take the maximum total system benefit as the objective function, and includes at least the power balance constraints between partitions and the operation constraints of hydropower stations within the partitions.

[0061] In this embodiment, maximizing the overall system benefit can be expressed as maximizing the total economic benefit of the entire network through optimized scheduling. For example, the objective function can be the net benefit obtained by subtracting the total electricity purchase cost of all input zones from the total electricity sales revenue of all output zones during the scheduling period. Specifically, this can be calculated by multiplying the net output (or net load) of each zone by the electricity price of the corresponding time period and then summing the results.

[0062] In this embodiment, the inter-regional power balance constraint can be expressed as a means to ensure that the power exchange between different regions meets the requirements for safe operation of the power grid. Specifically, it may require that during any scheduling period, the sum of the net power output of all output-type regions must be equal to the sum of the net power received by all input-type regions (i.e., real-time power balance of the entire network), while the transmission power of each interconnected transmission channel must not exceed its thermal stability limit or stability limit.

[0063] In this embodiment, the operational constraints of the hydropower stations within the designated area can be used to ensure the stable operation of each hydropower station under both structural safety and hydrological conditions. Specifically, these constraints may further include: water balance constraints (the change in reservoir water volume equals the difference between inflow and outflow), power generation flow constraints (the turbine's flow capacity is limited), downstream flow constraints (the minimum and maximum flow rates required to meet downstream ecological and flood control requirements), reservoir capacity constraints (the reservoir's operating water level is between the dead water level and the normal storage water level), power output constraints (the technical output range of the turbine generator units), water level fluctuation constraints (the limit on water level changes per unit time), and initial and final water level constraints (the set water level values ​​for the start and end of the scheduling period). These constraints collectively constitute the boundary conditions for the safe operation of the hydropower stations.

[0064] In this embodiment, the preset optimization algorithm can be a stepwise optimization algorithm. Specifically, a stepwise optimization algorithm can be used, which decomposes the multi-stage decision problem into a series of two-stage optimization sub-problems and avoids the curse of dimensionality through iterative solutions. More specifically, the scheduling period can be discretized first, and then, starting from the initial state, the time windows consisting of two adjacent time periods are optimized sequentially, while fixing the states of other time periods, gradually approaching the optimal solution until the objective function converges. After the solution is completed, the scheduling instructions such as the power generation plan of each hydropower station for each time period and the reservoir water level process line are output.

[0065] In this embodiment, the preset optimization algorithm can also be the Lagrange relaxation method. Specifically, by introducing Lagrange multipliers, the coupling constraints in the model that cause difficulty in solving (e.g., interval power balance constraints) can be relaxed and penalized before being incorporated into the objective function, thereby decomposing the original complex problem into several easily solvable subproblems (e.g., subproblems based on a single hydropower station or a single virtual scheduling partition). By iteratively updating the Lagrange multipliers, the solutions to these subproblems can be coordinated, and the optimal solution to the original problem can be gradually approximated.

[0066] In this embodiment, the preset optimization algorithm can also be a genetic algorithm. Specifically, it can simulate natural selection and genetic mechanisms, encode potential scheduling schemes as individuals of chromosomes, initialize a population, and iteratively perform operations such as selection, crossover (exchanging some scheduling instructions), and mutation (random micro-scheduling instructions). The fitness of individuals is evaluated based on the objective function value, thereby eliminating the inferior and guiding the population to evolve towards a better scheduling scheme.

[0067] In this embodiment, the preset optimization algorithm can also be a particle swarm optimization algorithm. Specifically, in this algorithm, each potential scheduling scheme is regarded as a particle in the search space. The particle swarm flies in the solution space, and each particle dynamically adjusts its flight direction and speed according to its own historical best position and the group's historical best position, that is, updates its scheduling scheme.

[0068] The method provided in this embodiment introduces a virtual scheduling partitioning mechanism to decompose a complex inter-basin hydropower station system into multiple collaboratively operating sub-units. First, it rationally divides the power station group based on the coupling relationship between hydraulics and electricity to significantly reduce the dimensionality of the optimization problem and effectively avoid the curse of dimensionality. Then, by dynamically determining the input / output attributes of each partition, it accurately identifies the real-time supply and demand characteristics and regulation potential of different regions. Finally, it performs collaborative optimization using partitions as the basic scheduling unit, maximizing the overall system operational efficiency while ensuring that various complex constraints are met. This achieves a simultaneous improvement in scheduling computation efficiency and the complementary benefits of inter-basin hydropower collaboration.

[0069] In some embodiments, the step of dividing the hydropower station group into multiple virtual dispatch zones based on the hydraulic and electrical interconnection strengths between the hydropower stations in the inter-basin hydropower station group includes:

[0070] Step S122: For any two hydropower stations, calculate the power connection strength index and hydraulic connection strength index between them respectively.

[0071] In this embodiment, the power linkage strength index can be determined based on calculations derived from the electrical characteristics analysis of the power grid. This power linkage strength index characterizes the power linkage strength between power plants by quantifying the degree of electrical coupling between nodes. Specifically, firstly, the topology parameters and operational data of the power grid can be obtained to establish a network model including all hydropower plant nodes. Then, the linkage strength is measured by calculating the equivalent electrical distance between nodes; the smaller the distance value, the higher the degree of electrical coupling. The calculation of the electrical distance can comprehensively consider the interactive effects of changes in active and reactive power on node voltages, and can be implemented using methods based on impedance matrices or sensitivity analysis.

[0072] In this embodiment, the hydraulic connection strength index can be determined based on calculations analyzing the physical hydraulic connections between hydropower stations. Specifically, this hydraulic connection strength index may include a flow lag factor and a reservoir capacity regulation factor. The flow lag factor reflects the time delay required for water to flow from the upstream hydropower station to the downstream hydropower station. The shorter the delay, the larger the value of this factor, indicating better synchronization of the hydraulic responses of the two power stations. The reservoir capacity regulation factor reflects the impact of the upstream hydropower station's reservoir regulation capacity on the operation of the downstream hydropower station. The larger the effective regulation capacity of the upstream reservoir (the difference between the maximum and minimum reservoir capacity), the larger the value of this factor, indicating a greater potential for coordinated operation of the two power stations through reservoir scheduling.

[0073] Step S124: Normalize the electrical connection strength index and the hydraulic connection strength index, and then weight and fuse them to obtain the comprehensive connection strength;

[0074] In this embodiment, since the power linkage strength index and the hydraulic linkage strength index have different dimensions and numerical distribution ranges, direct comparison or fusion would affect the rationality of the zoning results. Therefore, normalization processing is required to transform both types of indices into a dimensionless numerical range. Specifically, for the power linkage strength index, if a larger original value indicates a weaker linkage, a transformation method such as 1 - (current value / maximum value) can be used to ensure that a larger transformed value represents a stronger linkage. For the hydraulic linkage strength index, if its original value is already positively correlated with the linkage strength, a scaling method such as (current value - minimum value) / (maximum value - minimum value) can be used.

[0075] In this implementation, after normalization, a weighted fusion is required to obtain the overall connectivity strength. The weighting coefficients can be configured according to the characteristics of the actual system. Specifically, for systems where hydropower is the main means of regulation, the weight of hydraulic connectivity strength can be appropriately increased; for systems with complex power grid structures and prominent transmission constraints, the emphasis can be placed on power connectivity strength.

[0076] Step S126: Construct a complex network of hydropower stations with each hydropower station as a node and the comprehensive connection strength as the edge weight.

[0077] In this embodiment, each hydropower station in the river basin can be considered an independent node in the network. An edge can be established between any two hydropower station nodes, and the comprehensive connection strength value between the two stations calculated in step S124 is assigned to this edge as its weight. The network constructed in this way is a weighted undirected graph. That is, the larger the weight of the edge, the closer the connection between the corresponding two hydropower station nodes in the electricity-water comprehensive dimension, and the more they should be considered to be assigned to the same partition.

[0078] Step S128: Based on the principle of maximizing modularity, the complex network of the hydropower station is divided into communities, and the communities obtained are used as the virtual scheduling partitions.

[0079] In this embodiment, the modularity is used as a quantitative indicator to evaluate the quality of the network community structure. Specifically, it measures the difference between the actual weights of edges within a community under a given partition and the expected weights in a random network. The closer the modularity is to 1, the more obvious the community structure and the better the partitioning effect.

[0080] In this embodiment, the community partitioning of the complex hydropower station network based on the principle of maximizing modularity can be represented as using a preset community discovery algorithm to find a partitioning scheme that makes the modularity index of the entire network reach or approach its maximum value. This community discovery algorithm can be the Louvain algorithm. The Louvain algorithm iteratively optimizes and continuously adjusts node affiliations until the modularity reaches its maximum value. Specifically, each node can first be initialized as an independent community, and then iteratively attempts to move nodes to adjacent communities or merge communities. Each operation evaluates the change in modularity and accepts changes that increase modularity. When any movement or merging can no longer significantly improve modularity, the algorithm terminates, and the resulting stable community structure is the final virtual scheduling partitioning scheme.

[0081] In this embodiment, the community discovery algorithm can also be the GN algorithm. Specifically, firstly, the edge betweenness number (i.e., the number of paths passing through that edge in the shortest path for all node pairs) of all edges in the network is calculated. Then, the edge with the highest current edge betweenness number is removed, and the edge betweenness number of the remaining edges in the network is recalculated. This process is repeated until the network is decomposed into a predetermined number of communities or all edges have been removed.

[0082] In this embodiment, the community discovery algorithm can also be a label propagation algorithm. Specifically, during initialization, each node can be assigned a unique label (i.e., a community identifier). During iteration, each node updates its own label according to the labels appearing in its neighboring nodes, using a preset strategy (e.g., selecting the label with the highest frequency among its neighbors). Through multiple iterations, the labels will reach consensus within closely connected groups of nodes, thus naturally forming a community structure.

[0083] In some implementations, the step of calculating the power linkage strength index and hydraulic linkage strength index between any two hydropower stations includes:

[0084] Step S1222: Based on the sensitivity analysis of the power grid, determine the electrical distance reflecting the electrical coupling relationship between the nodes of each hydropower station, and use it as the power connection strength index; wherein, the electrical distance is negatively correlated with the power connection strength between the two hydropower stations.

[0085] In this embodiment, the sensitivity analysis of the power grid can be expressed as analyzing the response characteristics of power system state variables (e.g., node voltages) to changes in control variables (e.g., node injected power), that is, establishing a differential relationship matrix between node electrical parameters.

[0086] In one specific implementation, step S1222 may include:

[0087] Step S12221: Obtain basic power grid parameters and construct a mathematical model. Specifically, first, basic data such as the current power grid topology, line parameters (including resistance, reactance, and susceptance), and transformer turns ratios can be obtained from the power grid energy management system. Based on this data, a power grid mathematical model for power flow calculation can be established, which can be represented as a node admittance matrix. This matrix can be the core mathematical model describing the electrical connection relationships and characteristics between the nodes of the power grid.

[0088] Step S12222: Calculate the sensitivity matrix. Specifically, under a given power grid operating reference point (which can be the current or typical operating mode), perform power flow calculations. Based on the results of the power flow calculations and the established nodal admittance matrix, calculate the required voltage sensitivity matrix by obtaining partial derivatives or using the adjoint network method.

[0089] Step S12223: Construct the electrical distance component. Specifically, using the active power voltage sensitivity matrix obtained in sub-step S12222, an electrical distance component based on active power sensitivity can be constructed for any two hydropower station nodes i and j. Then, the Euclidean distance between these two row vectors is calculated as the initial electrical distance component between nodes i and j. Similarly, the electrical distance component based on reactive power sensitivity can be calculated using the reactive power voltage sensitivity matrix in the same way.

[0090] Step S12224: Calculate the comprehensive electrical distance. Specifically, the two electrical distance components calculated in sub-step S1222c can be fused to form a comprehensive electrical distance index that reflects the effects of active and reactive power. For example, this fusion method can be to take the arithmetic mean of the two components.

[0091] Step S12225: Determine the power connection strength index. Specifically, since the electrical distance and power connection strength are defined as negatively correlated, that is, the larger the distance value, the weaker the connection, the calculated comprehensive electrical distance value can be directly used as the final power connection strength index.

[0092] Step S1224: Based on the topological relationship between hydropower stations and the regulation characteristics of reservoirs, determine a comprehensive factor reflecting their hydraulic coupling relationship as the hydraulic connection strength index; wherein, the comprehensive factor includes at least a flow lag factor reflecting the water flow propagation time and a reservoir capacity regulation factor reflecting the reservoir regulation capacity.

[0093] In this embodiment, the topological relationship can be represented as the physical connection structure formed between hydropower stations through river channels, specifically including spatial layout features such as cascade connection relationships and tributary confluence relationships. In practical implementation, a hydrological network model of the hydropower station group can be established to clarify the upstream and downstream connection relationships and river parameters of each power station.

[0094] In this embodiment, a shorter water propagation time means a faster hydraulic response from the upstream and downstream power stations and a greater potential for coordinated operation. Therefore, the value of the water flow lag factor should be larger, i.e., it is negatively correlated with the water propagation time. Thus, the water flow lag factor can be defined as an inverse function or a negative exponential function of the water propagation time. In a specific implementation plan, the water propagation time can be determined using hydraulic formulas or historical hydrological data statistics based on parameters such as river length, gradient, and flow rate, and then the water flow lag factor can be calculated.

[0095] In this embodiment, the larger the effective regulating capacity of the upstream reservoir (which can be the difference between the maximum and minimum capacity), the stronger its ability to redistribute runoff and smooth downstream flow, and the greater the benefits that can be generated from coordinated scheduling with downstream power plants. Therefore, the value of this factor should be larger, meaning it is positively correlated with the effective regulating capacity of the upstream reservoir. Thus, this capacity regulating factor can be defined as the ratio of the effective regulating capacity of the upstream reservoir to a certain benchmark capacity of the downstream reservoir, or other related functions.

[0096] In this embodiment, the final hydraulic connection strength index can be a single value obtained by weighted summation or other function fusion methods of the above-mentioned flow lag factor and reservoir capacity adjustment factor.

[0097] In one specific implementation, step S1224 may include:

[0098] Step S12241: Determine the topological relationships of hydropower stations. Specifically, firstly, the topological connections between hydropower station groups can be determined based on the river basin map and the geographical location information of the hydropower stations. More specifically, an upstream and downstream relationship diagram of hydropower stations can be established to determine which hydropower stations have direct hydraulic connections (i.e., water flows directly from the upstream hydropower station to the downstream hydropower station).

[0099] Step S12242: Calculate the flow lag factor. Specifically, for any two hydropower stations i and j (i is upstream, j is downstream) with a direct upstream-downstream relationship, the flow lag factor between them can be calculated. More specifically, the flow propagation time from hydropower station i to hydropower station j can be determined. This time can be determined by hydraulic formulas (e.g., based on channel length, slope, roughness coefficient, and typical flow rate) or by fitting historical hydrological observation data. The flow propagation time can be converted into a flow lag factor, which can be constructed to be negatively correlated with the propagation time; that is, the longer the propagation time, the smaller the factor value, indicating a weaker hydraulic connection. For example, it can be defined as an inverse function of the propagation time.

[0100] Step S12243: Calculate the reservoir capacity regulation factor. Specifically, for any two hydropower stations i and j with a direct upstream-downstream relationship, the reservoir capacity regulation factor between them can be calculated. More specifically, the reservoir characteristic parameters of hydropower station i (upstream) and hydropower station j (downstream) can be obtained: the maximum and minimum reservoir capacities of the upstream reservoir, and the maximum and minimum reservoir capacities of the downstream reservoir. The effective regulating capacity of the upstream reservoir can be calculated; the effective regulating capacity of the downstream reservoir can be calculated; the reservoir capacity regulation factor can be calculated, which can be constructed to be positively correlated with the regulating capacity of the upstream reservoir. For example, the ratio of the effective regulating capacities of the upstream and downstream reservoirs can be calculated.

[0101] Step S12244: Obtain the hydraulic connection strength index through weighted fusion. Specifically, the calculated flow lag factor and reservoir capacity regulation factor can be weighted and fused to obtain the final hydraulic connection strength index.

[0102] In some implementations, the step of dividing the complex network of the hydropower station into communities based on the principle of maximizing modularity, and using the resulting communities as the virtual scheduling partitions, includes:

[0103] Step S1282: Using a heuristic algorithm based on modularity gain, iterative community detection is performed on the complex network of the hydropower station until the network modularity no longer increases, so as to obtain the virtual scheduling partition.

[0104] In this embodiment, the complex network of a hydropower station can be divided into a community structure (virtual scheduling partition) with tightly connected internal connections and relatively sparse external connections through a preset iterative optimization process.

[0105] In one specific implementation, step S1282 may include:

[0106] Step S12821: Perform network initialization and modularity benchmark calculation. Specifically, firstly, each hydropower station node in the complex hydropower station network can be initialized as an independent community; that is, in the initial state, there are communities in the network equal to the number of nodes. Then, the initial modularity value under the current network partition can be calculated, and this value can be used as the benchmark for modularity optimization. Modularity is an indicator that measures the quality of network community partitioning, and its calculation can be based on the overall edge weight structure of the network and the current community allocation scheme.

[0107] Step S12822: Perform local optimization and modularity gain evaluation. Specifically, an iterative loop can be entered to traverse the nodes in the network. For the currently selected node, it can be removed from its current community, and the possibility of merging it into the community of one of its neighboring nodes can be examined one by one. For each possible community transfer, the modularity gain resulting from this action is calculated, which is the difference in the overall modularity of the network before and after the transfer. The modularity gain quantifies the degree to which this community change improves the strength of the community structure.

[0108] Step S12823: Perform community assignment update. Specifically, after evaluating all possible transfer options, the algorithm selects the community transfer scheme that yields the maximum positive modularity gain. If such a scheme exists (i.e., the maximum modularity gain is greater than zero), the community change is executed, and the current node is officially assigned to its corresponding new community. If none of the possible transfers produce a positive modularity gain, the node's current community affiliation remains unchanged. This process traverses the nodes in the network once, completing one local optimization.

[0109] Step S12824: Network Reconstruction and Iteration. Specifically, after completing one round of local optimization for all nodes, each currently defined community is treated as a new supernode, and a new, scaled-down network is reconstructed based on the original connection relationships. In the new network, the edge weights between supernodes are the sum of the weights of all edges between corresponding communities in the original network, while the edge weights within a supernode are aggregated into self-loop weights. On this reconstructed new network, the local optimization and community allocation processes of steps S12822 and S12823 are repeated.

[0110] Step S12825: Termination Judgment and Result Output. Specifically, repeat network reconstruction and iterative optimization until, after a certain iteration, the network's community structure no longer changes or the improvement in modularity falls below a preset small threshold. At this point, the algorithm terminates and outputs the final stable community partitioning result. Each community in this result is defined as a virtual scheduling partition.

[0111] In some implementations, the step of determining whether a virtual scheduling partition is an input partition or an output partition based on the partition's power supply and demand balance conditions and the hydropower station's end-of-day water level constraints includes:

[0112] If the total real-time output of a virtual dispatching zone during the dispatching period is greater than or equal to its load demand, and the water level at the end of the dispatching period of each hydropower station in the zone is not lower than its preset end-of-day water level boundary condition, then the zone is determined to be an output-type zone.

[0113] In this embodiment, a region can be classified as an output-type region if both the power supply and demand balance condition and the water level constraint condition are met simultaneously. Specifically, the power supply and demand balance condition can be expressed as follows: in each calculation period within the dispatch cycle, the real-time total output of the region (i.e., the sum of the outputs of all hydropower stations, wind power stations, and photovoltaic power stations within the region during that period) is greater than or equal to the load demand of the region during that period. The water level constraint condition can be expressed as follows: at the end of the dispatch cycle, the actual reservoir water level of each hydropower station within the region is not lower than its respective preset end-of-day water level boundary condition. Meeting this condition means that the region is not only completely self-sufficient throughout the dispatch period but also has sufficient power generation margin, and can still maintain a good reservoir storage state at the end of the dispatch period, possessing the ability to transmit power to other regions.

[0114] If the total real-time output of a virtual dispatching zone during the dispatching period is less than its load demand, or if the water level at the end of the dispatching period of any hydropower station in the zone is lower than its preset end-of-day water level boundary condition in order to meet the load demand, then the zone is determined to be an input-type zone.

[0115] In this embodiment, a zone is classified as an input-type zone if either the power shortage condition or the water level exceeding the limit condition is met. Specifically, the power shortage condition can be expressed as follows: during any calculation period within the scheduling cycle, the zone's real-time total output is less than its load demand for that period. This indicates that the zone's own power generation capacity is insufficient to meet its internal load and it must rely on external power input. The water level exceeding the limit condition can be expressed as follows: although the zone's real-time total output is adjusted to equal its load demand in each calculation period within the scheduling cycle by adjusting hydropower output, achieving this power balance results in at least one hydropower station within the zone having an actual reservoir water level lower than its preset end-of-day water level boundary condition at the end of the scheduling cycle.

[0116] In a specific implementation plan, the step of determining whether a virtual scheduling partition is an input partition or an output partition based on the power supply and demand balance conditions and the end-of-day water level constraints of the hydropower station for each virtual scheduling partition may include:

[0117] Step S142: Load demand data for each virtual dispatching zone throughout the entire dispatching cycle can be obtained from the power grid energy management system. Simultaneously, initial water levels, inflow forecasts, and output predictions for wind and solar power stations within the zone can be obtained from the reservoir dispatching system or meteorological forecasting system. Then, the day-end water level boundary conditions for each hydropower station are retrieved from the database.

[0118] Step S144: The calculation of the regional power balance simulation can be performed. Specifically, for each virtual scheduling region, the controller can simulate the operation of that region within the scheduling cycle based on the data obtained in step S142. More specifically, under the premise of satisfying the inherent basic operational constraints of each hydropower station within the region, such as hydraulics, reservoir capacity, and output, a preliminary scheduling calculation is performed with the goal of maximizing the satisfaction of the region's load demand. This calculation outputs the planned output of each power source within the region for each time period and the simulated end-of-period water level of each hydropower station.

[0119] Step S146: Perform a check of the judgment conditions. That is, based on the simulation results of step S144, perform a check of the power supply and demand balance condition and the water level constraint satisfaction condition. If both conditions are met simultaneously, the partition is judged as an output type partition. If the partition fails the above two conditions, the controller performs a second judgment check, that is, a judgment of the power shortage condition or the water level exceeding the limit condition. If either condition is met, the partition is judged as an input type partition.

[0120] Step S148: Based on the judgment result, the controller assigns output or input attribute labels to the currently processed virtual scheduling partition.

[0121] In some implementations, the objective function of the preset short-term optimization scheduling model is configured as follows:

[0122] Maximize the net difference between the total electricity sales revenue of all output-type zones and the total electricity purchase cost of all input-type zones during the dispatch period;

[0123] The pre-defined short-term optimization scheduling model's inter-regional power balance constraint is configured such that the sum of the net output of all output-type zones during any scheduling period is equal to the sum of the net load demand of all input-type zones during the same period.

[0124] In this embodiment, for each partition identified as an output-type partition, during each time period of the scheduling period, its net output available for external transmission (i.e., the portion of the partition's total output exceeding its own load demand) is multiplied by the partition's electricity sales price for that time period to obtain the partition's electricity sales revenue for that time period. The electricity sales revenue of all output-type partitions across all time periods is summed. For each partition identified as an input-type partition, during each time period of the scheduling period, its net load demand that needs to be imported from the outside (i.e., the portion of the partition's load demand exceeding its own total output) is multiplied by the partition's electricity purchase price for that time period to obtain the partition's electricity purchase cost for that time period. The electricity purchase costs of all input-type partitions across all time periods are summed. Finally, the total electricity sales revenue is subtracted from the total electricity purchase cost to obtain the total net system benefit during the scheduling period. The objective function is then configured to seek the scheduling scheme that maximizes this net benefit value.

[0125] The operational constraints of the hydropower stations within the specified zone are configured as follows: water balance constraints, power generation flow constraints, discharge flow constraints, reservoir capacity constraints, hydraulic connection constraints, power output constraints, water level fluctuation constraints, and initial and final water level constraints.

[0126] In this embodiment, the water balance constraint can be expressed as requiring the reservoir capacity of the hydropower station at the end of any time period to be equal to its initial reservoir capacity at the beginning of the time period plus the product of the difference between the inflow and outflow during that time period and the length of the time period. This constraint ensures the continuity of water volume.

[0127] The power generation flow constraint can be expressed as the requirement that the power generation flow referenced by the hydropower station unit at any time period shall not exceed the technical upper limit determined by the maximum flow capacity of its turbine.

[0128] The discharge constraint can be expressed as a requirement that the discharge flow of a hydropower station reservoir at any given time be limited to a minimum and a maximum value. The lower limit can be determined by downstream ecological, water supply, or navigation needs, while the upper limit is determined by downstream flood control safety requirements.

[0129] Reservoir capacity constraints can be expressed as requiring that the reservoir capacity (or corresponding water level) of a hydropower station at any given time must not be lower than the dead capacity (dead water level) and must not be higher than the maximum allowable capacity determined by the normal storage level or flood control limit level.

[0130] The hydraulic connection constraint can be expressed as follows: for cascade hydropower stations with direct hydraulic connections, the inflow of the downstream power station in any period is equal to the sum of the outflows of all its directly upstream power stations during the same period, plus the inflow between the two stations.

[0131] Output constraints can be expressed as the requirement that the output of a hydropower station unit at any given time be limited between its minimum technical output and its rated capacity.

[0132] Water level fluctuation constraints can be expressed as the requirement that the water level fluctuation (rate of rise or fall) of a hydropower station reservoir at any given time must not exceed the maximum allowable fluctuation set to ensure the stability of the dam structure and reservoir banks.

[0133] Initial and final water level constraints can be expressed as requiring the water level of the hydropower station reservoir at the beginning of the scheduling period to be equal to the specified initial water level (e.g., normal storage water level), and the water level at the end of the scheduling period to be equal to the specified target water level (e.g., end-of-day water level boundary condition or end-of-cycle control water level).

[0134] In some implementations, the step of using the virtual scheduling partition as the scheduling unit and solving a preset short-term optimization scheduling model with a preset optimization algorithm to obtain the short-term scheduling plan for the cross-basin hydropower station group includes:

[0135] Step S162: Solve the short-term optimization scheduling model using a stepwise optimization algorithm; wherein, the multi-stage decision problem of the entire scheduling period is decomposed into a series of sequentially executed two-stage optimization sub-problems, and each two-stage sub-problem is optimized and solved, and the short-term scheduling plan is obtained through iterative calculation.

[0136] In this embodiment, step S162 may include:

[0137] Step S1621: Perform scheduling cycle discretization and initial trajectory generation. Specifically, first, the entire scheduling cycle (e.g., 24 hours) is uniformly discretized into several time periods (e.g., 96 15-minute time periods). Then, an initial state trajectory (which can be a reservoir water level process line) is generated for each hydropower station throughout the entire scheduling cycle. This initial trajectory can be generated based on historical operating data, a constant water level strategy, or other heuristic methods.

[0138] Step S1622: Set the iterative process and convergence criteria. Specifically, start the iterative solution loop. Convergence criteria can be preset, such as the improvement in the objective function value between two adjacent iterations being less than a specified small positive number, or reaching the maximum number of iterations.

[0139] Step S1623: Sequentially construct and solve the two-stage subproblems. Specifically, in each iteration, the two-stage optimization subproblems consisting of two adjacent time periods (denoted as time period j and time period j+1) are processed sequentially in chronological order (from the first time period to the last time period). For each subproblem, the following data processing is performed:

[0140] First, except for the current processing time periods j and j+1, the state of the hydropower station (mainly the reservoir water level) in all other time periods within the scheduling cycle is fixed to the result obtained in the previous iteration (or the value of the initial trajectory).

[0141] Then, given that the state of time period j is known and the state of time period j+2 and beyond is fixed, the decision variables for time periods j and j+1 (e.g., the output and discharge flow of each hydropower station) are used as optimization variables to construct a local optimization problem that applies only to these two time periods. The objective function of this subproblem is the part of the global objective function that is relevant to these two time periods, and its constraints include all interval and intra-regional constraints relevant to these two time periods.

[0142] Finally, a pre-defined planning algorithm (e.g., a sequential quadratic programming algorithm or interior point method for nonlinear problems) is invoked to solve the two-stage optimization subproblem with a significantly reduced scale, obtain the local optimal decisions for time periods j and j+1, and update the final state (reservoir water level) for time period j+1.

[0143] Step S1624: Update the state trajectory and advance window. Specifically, after solving a two-phase subproblem, the old value is replaced with the new decision and the updated state (especially the final water level of time period j+1), thereby updating the corresponding time period portion of the entire state trajectory. Then, the time window is moved forward by one time period (i.e., processing time periods j+1 and j+2), and the process of step S1623 is repeated until all adjacent time period pairs within the scheduling cycle have been processed.

[0144] Step S1625: After completing a full round of sequential optimization for all adjacent time periods, calculate the global objective function value corresponding to the complete scheduling plan under the current iteration. Check whether the improvement of this value relative to the previous iteration meets the preset convergence criterion. If the convergence criterion is not met, restart the next iteration based on the complete state trajectory updated in this iteration, i.e., jump to step S1623, and optimize all two-stage sub-problems sequentially again. If the convergence criterion is met, the iteration terminates. At this point, the time-period output plan of each hydropower station, the discharge flow process, and the reservoir water level process line obtained from the last iteration constitute the final short-term scheduling plan for the inter-basin hydropower station group, which is output by the controller for execution.

[0145] In one specific implementation plan, a short-term optimal scheduling method for cross-basin hydropower station groups considering virtual scheduling partitions is provided.

[0146] In summary, existing research largely focuses on the scheduling optimization of cascade hydropower stations in a single river basin, lacking a systematic scheduling framework that can coordinate the entire chain of "zoning-assessment-modeling," making it difficult to fully leverage the scale advantages and complementary benefits of inter-basin hydropower station clusters. To address this, this invention introduces the core concept of "virtual scheduling zoning," creatively decomposing the complex inter-basin hydropower station cluster scheduling problem into three levels: "zoning division," "zoning discrimination," and "zoning coordination." Through scientific division, attribute discrimination, and the application of zoned coordinated scheduling, it systematically solves the pain points of existing technologies, achieving a significant improvement in scheduling efficiency.

[0147] To address the challenges of managing numerous power stations brought about by the rapid increase in the number and scale of power plants, this invention proposes a short-term optimal scheduling method for inter-basin hydropower station groups considering virtual scheduling partitions. First, the hydropower station group is divided into virtual scheduling partitions based on the hydraulic-electrical connections between stations. Then, based on supply and demand balance analysis within each partition, the virtual scheduling partitions are further divided into input-type and output-type partitions. Finally, a short-term scheduling model is constructed using the virtual scheduling partitions as scheduling units, aiming to maximize residual energy storage, thereby achieving efficient scheduling of large-scale inter-basin hydropower station groups. The results of the first step, virtual scheduling partitioning, form the basis for the second step's partition attribute determination. The attribute determination results of the second step directly define the "role" of each partition in the third step's optimization model, simplifying the complex power balance constraints of the entire network nodes into directional power exchange constraints based on partitions. This method, using virtual scheduling partitions as the basic unit for modeling, significantly reduces the dimensionality of decision variables, transforms complex grid security constraints into key transmission section constraints, and improves model solution efficiency.

[0148] 1.1 Virtual Dispatch Partitioning under Complex Hydraulic-Electric Coupling Constraints

[0149] 1.1.1 Virtual Scheduling Partitioning Indicators

[0150] A well-defined zoning of hydropower station groups can reduce the number of system dispatching units and improve dispatching and management efficiency. When dividing inter-basin hydropower station groups into zones, in addition to considering the electrical distances between stations, the hydraulic connections between them should also be emphasized. Grouping hydropower stations with strong hydraulic connections into the same zone ensures the effective utilization of flexible hydropower regulation capabilities.

[0151] (1) Power connection indicators

[0152] When virtually scheduling and partitioning a group of hydropower stations across a river basin, priority should be given to the electrical connections between the stations. Stations with strong electrical connections are typically grouped into the same partition to improve information transmission efficiency and minimize power transmission losses. The strength of these electrical connections is usually quantified by the electrical distance between stations, as defined below:

[0153] (1);

[0154] (2);

[0155] (3);

[0156] (4);

[0157] (5);

[0158] (6);

[0159] (7);

[0160] In the formula, This represents the active-voltage electrical distance between node i and node j. This value is obtained by calculating the Euclidean distance of the voltage-active sensitivity ratio vector of the two nodes to all other nodes (1 to n). The larger this distance value, the weaker the electrical connection between the two nodes in the active-voltage dimension. This represents the active power-voltage sensitivity matrix of the power grid. This matrix describes the relationship between changes in the active power injected into nodes and the node voltage amplitude. This represents the vector of changes in the active power injected into the node. This represents the vector of changes in node voltage magnitude caused by changes in active power. This can be expressed as a ratio parameter, which is the logarithm of the ratio of the voltage change at node j to the voltage change at node i when a unit change in injected active power occurs. The larger this ratio, the greater the impact of the power change at node j on itself compared to its impact on node i, indicating a weaker electrical coupling and a greater electrical distance between the two nodes. This represents the reactive-voltage electrical distance between node i and node j. This represents the reactive power-voltage sensitivity matrix of the power grid. This matrix describes the relationship between changes in reactive power injected into nodes and the magnitude of node voltage. This represents the vector of changes in node voltage magnitude caused by changes in reactive power. This represents the vector of changes in reactive power injected into the node. Is with The corresponding ratio parameter is used to measure the degree of electrical isolation between two nodes under reactive power disturbance. This represents the comprehensive electrical distance between node i and node j. This parameter is the final metric after comprehensively considering the effects of active and reactive power.

[0161] (2) Hydraulic linkage index

[0162] For power supply systems that rely on hydropower as the core regulator, the hydraulic connectivity between hydropower stations is a key factor to consider when creating virtual zones. Hydropower stations with strong hydraulic connectivity should be grouped into the same zone to leverage their collaborative operation advantages. The strength of the hydraulic connectivity between hydropower stations is primarily assessed through the following two aspects:

[0163] 1) Water flow lag factor

[0164] A key factor influencing the strength of hydraulic connection in a hydropower station is the flow delay between upstream and downstream cascade stations. Shorter flow delays between upstream and downstream stations indicate stronger hydraulic connection, and vice versa. The corresponding functional expression is as follows:

[0165] (8);

[0166] In the formula: This factor is used to quantify the intensity of the time dimension of hydraulic connections, characterizing the speed at which water flows from an upstream hydropower station (i) to its downstream hydropower station (j). A larger factor value indicates that the water release from the upstream station affects the downstream station more quickly, suggesting a better hydraulic time-response and a closer connection between the two. Conversely, a smaller factor value indicates a slower and weaker hydraulic connection over time. This represents the actual propagation time required for water to flow from the upstream hydropower station (i) to the downstream hydropower station (j). This is an empirical coefficient used to account for the nonlinear effects of flow attenuation or variation along the river due to factors such as evaporation, seepage, and tributary inflow as water propagates through the river.

[0167] 2) Storage capacity adjustment factor

[0168] The synergistic operation benefits between upstream and downstream power plants largely depend on the regulation capacity of the upstream reservoir. A large upstream reservoir regulation capacity can redistribute natural runoff, generating significant synergistic benefits. Conversely, when the upstream reservoir regulation capacity is limited, the power plant typically operates as a runoff power plant, resulting in relatively smaller synergistic operation benefits with the downstream power plant. To address this, this invention uses the following functional formula to characterize the influence of upstream and downstream reservoir capacity on the strength of the hydraulic connection between power plants.

[0169] (9);

[0170] In the formula: As an indicator used to quantify the strength of the hydraulic linkage capacity dimension, it characterizes the regulation capacity advantage of upstream reservoirs relative to downstream reservoirs. These represent the upper limit of the reservoir capacity (corresponding to the normal water level) that the upstream hydropower station (i) reservoir can reach during normal operation and the lower limit of the reservoir capacity (corresponding to the dead water level) that must be maintained. These represent the upper and lower limits of the reservoir capacity of the downstream hydropower station (j), respectively.

[0171] The hydraulic connection strength between hydropower stations should comprehensively consider the two factors mentioned above. The specific function expression is as follows:

[0172] (10);

[0173] (11);

[0174] In the formula: Indicates the strength of the hydraulic connection between hydropower stations; , These are the weighting coefficients of the water flow lag factor and the reservoir capacity adjustment factor, respectively.

[0175] 1.1.2 Construction of Community Network for Hydropower Station Clusters

[0176] Community structure and its detection methods are an important branch of complex network theory. They have been widely applied in sociology, business behavior analysis, and internet research, and are increasingly being used to study complex behaviors in power grids. Community structure is characterized by tightly interconnected nodes within a group and sparser connections between different groups, thus dividing the network into several tightly integrated blocks. Virtual scheduling partitioning of inter-basin hydropower stations exhibits similar characteristics to community structure in networks. Therefore, community detection algorithms can be used to achieve virtual partitioning of inter-basin hydropower station groups.

[0177] (1) Indicator normalization

[0178] Since the different trends of the virtual partitioning indices established in Section 2.1.1 represent different meanings—for example, a larger electrical distance indicates a weaker electrical connection between power stations, while a larger hydraulic connection index indicates a stronger hydraulic coupling strength—it is necessary to normalize each index before virtual partitioning to ensure that their values ​​fall within the range of [0,1], and that the closer they are to 1, the stronger the connection. Equation (12) is used to normalize the electrical connection index, and Equation (13) is used to normalize the hydraulic connection index.

[0179] (12);

[0180] (13);

[0181] In the formula: This represents the normalized power connection strength index between hydropower stations i and j. This represents the original electrical distance between hydropower stations i and j. This represents the total number of hydropower stations in the system. This represents the normalized hydraulic connection strength index between hydropower stations i and j. This represents the initial hydraulic connection strength index between hydropower stations i and j. and Represent the original hydraulic connection indices for all hydropower station pairings (i,j). The maximum and minimum values ​​in the range.

[0182] (2) Construction of inter-basin hydropower station network

[0183] A cross-basin hydropower station network is constructed by treating individual hydropower stations as nodes and the connection strength between them as edges. This study uses the following expression to calculate the connection strength between power stations.

[0184] (14);

[0185] (15);

[0186] In the formula: This represents the overall connection strength between hydropower stations i and j, and is the edge weight in constructing the complex network of hydropower stations. and This represents the calculated normalized power linkage strength index and the normalized hydraulic linkage strength index. , These represent the weighting coefficients of electrical connection strength and hydraulic connection strength in the comprehensive calculation, respectively.

[0187] 1.1.3 Virtual Scheduling Partition Solution Method

[0188] Modularity is commonly used in community detection during partitioning of complex networks to measure the strength of community structures. A modularity close to 1 indicates a strong community structure and better virtual partitioning performance; a modularity close to 0 or a negative value indicates a weak community structure and poorer virtual partitioning performance. The modularity calculation function is as follows:

[0189] (16);

[0190] In the formula: Indicates the modularity of the virtual partition of the power plant; Indicates the connection node and nodes Edge weights; It is the sum of the weights of all edges in the network; Indicates connection to node The sum of the weights of all edges; if the node and Located in the same partition ,otherwise , This is a community indicator function.

[0191] Partitioning a multi-basin hydropower station group is a high-dimensional combinatorial optimization problem. Traditional intelligent algorithms, such as genetic algorithms and particle swarm optimization, often face challenges such as the need for efficient initial solutions or the tendency to get trapped in local optima. To overcome these limitations, this invention employs a Louvain-based community structure algorithm to solve the virtual partitioning of a multi-basin hydropower station group. The basic principle of this algorithm is to first treat each individual node as a separate partition, and then merge partitions in the direction of increasing modularity until the modularity value reaches its maximum. The specific solution steps are as follows:

[0192] Step 1: Input basic parameters and calculate power connection and hydraulic connection indicators;

[0193] Step 2: Normalize the above indicators and construct the cross-basin hydropower station group network structure;

[0194] Step 3: Treat each node in the network as a separate partition, forming the first-level partitioning result, and calculate the initial modularity value of the network. ;

[0195] Step 4: Randomly select adjacent nodes i and j to merge them into a new virtual partition, and calculate the module degree increment. Then let .if Add node i to The corresponding partition; otherwise, keep the existing partition unchanged;

[0196] Step 5: Treat the determined partitions as new nodes and rebuild the power plant network structure; repeat step 4 until the modularity no longer increases, and then output the final result.

[0197] 2.2 Virtual Scheduling Partition Attribute Determination

[0198] The main objective of power system optimization scheduling is to ensure the real-time balance between power supply and grid demand. The power sources of the existing power system mainly include thermal power, hydropower, wind power, and photovoltaic power. Thermal power usually bears the base load, so it can be deducted in the supply and demand balance analysis. Only the remaining load relationship between the hydropower, wind power and photovoltaic power stations and the grid after deducting the output of thermal power is analyzed. Since the output of wind power and photovoltaic power is not adjustable, each virtual scheduling zone can only ensure the reliability of power supply in the zone by adjusting the output of hydropower stations under a given load. In addition to power generation, hydropower stations also undertake the corresponding comprehensive utilization needs such as flood control, navigation and irrigation. Therefore, each basin control center usually has relevant requirements for the end-of-day water level of each hydropower station. In this regard, this invention will use whether the end-of-day water level of each hydropower station in the zone meets the boundary condition requirements as the standard for classifying virtual scheduling zones. Specifically, when the power station in the virtual scheduling zone can guarantee the grid load demand in real time and the end-of-day water level of the hydropower station is higher than the end-of-day water level boundary (as in equation (17)), the zone is an output zone. In addition to ensuring its own power supply demand, the output zone also has a certain power generation capacity to support the power supply of other zones. When the power stations within a virtual dispatch zone cannot meet the power supply demand even at their maximum generating capacity (as shown in equation (18)), or when the water level of the hydropower stations in the zone is below the water level boundary condition at the end of the day to meet the power supply demand (as shown in equation (19)), the zone is an input zone. Input zones require compensation from power stations in other zones to ensure the reliability of power supply within the zone. The function expressions for the above input and output zones are as follows:

[0199] (17);

[0200] (18);

[0201] (19);

[0202] In the formula: c is the index of the virtual scheduling partition, t is the index of the scheduling period, and T is the end time of the scheduling period. They are respectively in the time period partition The power output of internal hydropower stations, wind power stations, and photovoltaic power stations; For the power grid during the time period For partitions Load demand issued by inland waterway wind and solar power stations (after deducting the load of thermal power and other power sources). Partitions at the end of the scheduling period Inland hydroelectric power station The reservoir water level and water level boundary conditions, m; Indexes for hydropower stations, wind power stations, and photovoltaic power stations, respectively; The numbers represent the total number of hydropower stations, wind power stations, and photovoltaic power stations within the zone, respectively. Formula (17) defines the conditions for classifying a zone as an output zone (power balance condition + sustainable water level condition). If both conditions are met, the zone is classified as an output zone. Formula (18) defines one scenario for classifying a zone as an input zone, which is that the power deficit condition is met. Even if all clean power sources in the zone generate electricity at their capacity, the total output still cannot meet the load demand, resulting in a clear power deficit. Once this occurs, regardless of the water level condition, the zone is classified as an input zone (power deficit requires power reception). Formula (19) defines another scenario for classifying a zone as an input zone, which is that the current power balance is achieved at the expense of reservoir safety and future regulation capacity, and is therefore unsustainable. Thus, the zone also needs to be classified as an input zone, requiring external power reception to reduce the output of hydropower stations to restore the water level.

[0203] 2.3 Cooperative Optimization Scheduling Model Based on Virtual Scheduling Partitions

[0204] 2.3.1 Objective Function

[0205] Once the attributes of the virtual scheduling partition are determined, the virtual partition can be simplified into a scheduling unit, optimizing the scheduling process of each power station. This invention aims to maximize the overall system benefit, and the specific functional expression is as follows:

[0206] (20);

[0207] In the formula: The total power generation benefit during the entire dispatch period; This represents the total number of time periods during the scheduling period; These represent the total number of output partitions and input partitions, respectively. For output partitioning exist Net output at that time; For input type partition exist Net load requirement at that time; Partitions , exist Electricity price at that time; , , These are the input partition, output partition, and scheduling period indexes, respectively. This refers to the duration of the scheduling period.

[0208] 2.3.2 Constraints

[0209] In addition to satisfying power supply balance and channel capacity constraints between zones, inter-zone coordinated scheduling must also meet relevant constraints on hydropower scheduling within each zone. These constraints are as follows:

[0210] (1) Interval constraints

[0211] 1) Constraints on power supply and demand balance:

[0212] (twenty one);

[0213] In the formula: The set of all partitions identified as power output type; The set of all partitions identified as power input type; For output partitioning exist Net output at that time For input type partition exist Net output at that time.

[0214] 2) External transmission channel capacity constraints:

[0215] (twenty two);

[0216] In the formula: For the first Each channel's transmission power; For the first Maximum allowed transmission power per channel This is the set of main power transmission channels connecting different virtual dispatching zones.

[0217] (2) Intra-partition constraints

[0218] 1) Water balance constraints:

[0219] (twenty three);

[0220] In the formula: , These represent the final reservoir capacity of the f-th level hydropower station in time periods j+1 and j, respectively. The discharge flow of the f-th level hydropower station in the j-th time period; This represents the inflow to the reservoir during the j-th time period for the f-th level hydropower station.

[0221] 2) Power generation flow constraints:

[0222] (twenty four);

[0223] In the formula: The maximum allowable power generation flow rate for the f-th level hydropower station is m³ / s.

[0224] 3) Discharge Constraints. To ensure the safety of downstream flood-prone areas and the ecological environment, discharge constraints must be met:

[0225] (25);

[0226] In the formula: This represents the minimum ecological water demand downstream of the f-th level hydropower station; This represents the maximum permissible discharge flow for the f-th level hydropower station.

[0227] 4) Reservoir capacity constraints of hydropower stations:

[0228] (26);

[0229] In the formula: This represents the dead storage capacity of the f-th level reservoir. This represents the maximum allowable storage capacity of the f-th level reservoir. During the flood season, it represents the storage capacity corresponding to the flood control limit water level, while at other times, it represents the storage capacity corresponding to the normal storage water level.

[0230] 5) Hydraulic constraints:

[0231] (27);

[0232] In the formula: This refers to the inflow rate of the (f+1)th level hydropower station during the j-th time period. This refers to the runoff during the j-th time period of the (f+1)th level hydropower station. This refers to the collection of upstream reservoirs that have a direct hydraulic relationship with the f+1 level hydropower station.

[0233] 6) Power output constraints of hydropower stations:

[0234] (28);

[0235] In the formula: , These are the minimum and maximum output constraints for the f-th level hydropower station.

[0236] 7) Reservoir water level fluctuation constraints:

[0237] (29);

[0238] In the formula: The water level change of the reservoir during the j-th time period; This represents the maximum permissible water level fluctuation of the reservoir during this period.

[0239] 8) Initial and final water level constraints:

[0240] (30);

[0241] In the formula: , These represent the initial and final water levels during the scheduling period of the f-th level reservoir. Water level boundary constraints.

[0242] 2.3.3 Solution Method

[0243] Since the relationship between residual energy storage and reservoir capacity / head is nonlinear, solving this model is a high-dimensional, nonlinear, and strongly time-dependent problem. Faced with this complex optimization problem—non-convex, high-dimensional, and multi-stage—conventional dynamic programming (DP) algorithms tend to grow exponentially with the number of state variables, easily encountering the "curse of dimensionality." Nonlinear programming solutions to non-convex problems are prone to getting trapped in local optima. Genetic or particle swarm optimization algorithms have high requirements for initial solutions and are also prone to getting trapped in local optima. However, the Progressive Optimization Algorithm (POA) is an intelligent variant of DP, its core idea being to decompose the multi-stage decision problem into a series of two-stage sub-problems. It can directly handle nonlinear models without linearization simplification, maintaining the physical reality and accuracy of the model, and has lower requirements for initial solutions. Based on these advantages of effectively balancing computational feasibility, model accuracy, and solution quality, this invention employs the POA algorithm to solve a short-term collaborative optimization scheduling model for a cross-basin hydropower station group based on virtual scheduling partitions.

[0244] This implementation method utilizes a virtual scheduling and partitioning method that considers the hydraulic-electricity connection to achieve a reasonable partitioning of cross-basin hydropower station groups. For example... Figure 3 As shown, cascade hydropower stations with strong hydraulic connections are grouped into the same zone and share the same power transmission channel. Similarly, inter-basin hydropower stations with strong power connections are also grouped into the same zone, facilitating centralized dispatch and management in the future. By using "virtual partitioning," high-dimensional complex problems are decomposed into multiple low-dimensional sub-problems, effectively avoiding the "curse of dimensionality." For example... Figure 4 As shown, it can be seen that compared with the solution efficiency of the unpartitioned solution, the solution speed of the present invention is significantly faster, meeting the real-time requirements of short-term scheduling.

[0245] According to an embodiment of the present invention, an electronic device is provided. The electronic device in this embodiment may include one or more of the following components: a processor, a network interface, memory, non-volatile memory, and one or more application programs, wherein the one or more application programs may be stored in the non-volatile memory and configured to be executed by one or more processors, and the one or more programs are configured to perform the methods described in the foregoing method embodiments.

[0246] According to embodiments of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a computer, causes the computer to perform the method described in any of the above embodiments.

[0247] According to embodiments of the present invention, a computer program product comprising instructions is also provided, which, when executed by a computer, cause the computer to perform a method in any of the above embodiments.

[0248] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A short-term optimal scheduling method for cross-basin hydropower station groups considering virtual scheduling partitions, characterized in that, Includes the following steps: Based on the strength of hydraulic and electrical connections between hydropower stations in the cross-basin hydropower station group, the hydropower station group is divided into multiple virtual scheduling zones. For each virtual dispatching partition, based on the power supply and demand balance conditions of the partition and the end-of-day water level constraints of the hydropower station, the virtual dispatching partition is identified as an input partition or an output partition; wherein, the input partition indicates a power shortage that requires power reception, and the output partition indicates a power surplus that can be transmitted to other regions; Using the virtual scheduling partition as the scheduling unit, a preset optimization algorithm is used to solve the preset short-term optimization scheduling model to obtain the short-term scheduling plan of the cross-basin hydropower station group; wherein, the short-term optimization scheduling model is configured to take the maximization of the total system benefit as the objective function, and includes at least the inter-regional power balance constraints and the hydropower station operation constraints within the partition.

2. The method according to claim 1, characterized in that, The step of dividing the hydropower station group into multiple virtual scheduling zones based on the hydraulic and electrical interconnection strengths between the hydropower stations in the inter-basin hydropower station group includes: For any two hydropower stations, calculate the power connection strength index and the hydraulic connection strength index between them respectively; The electrical connection strength index and the hydraulic connection strength index are normalized and then weighted and fused to obtain the comprehensive connection strength. A complex network of hydropower stations is constructed, with each hydropower station as a node and the comprehensive connection strength as the edge weight. Based on the principle of maximizing modularity, the complex network of the hydropower station is divided into communities, and the resulting communities are used as the virtual scheduling partitions.

3. The method according to claim 2, characterized in that, The steps for calculating the power linkage strength index and hydraulic linkage strength index between any two hydropower stations include: Based on the sensitivity analysis of the power grid, the electrical distance reflecting the electrical coupling relationship between the nodes of each hydropower station is determined as an indicator of the power connection strength; wherein, the electrical distance is negatively correlated with the power connection strength between two hydropower stations. Based on the topological relationship between hydropower stations and the regulation characteristics of reservoirs, a comprehensive factor reflecting their hydraulic coupling relationship is determined as an index of hydraulic connection strength; wherein, the comprehensive factor includes at least a flow lag factor reflecting the water flow propagation time and a reservoir capacity regulation factor reflecting the reservoir regulation capacity.

4. The method according to claim 3, characterized in that, The step of dividing the complex network of the hydropower station into communities based on the principle of maximizing modularity, and using the resulting communities as the virtual scheduling partitions, includes: A heuristic algorithm based on modularity gain is used to iteratively detect communities in the complex network of the hydropower station until the network modularity no longer increases, so as to obtain the virtual scheduling partition.

5. The method according to claim 1, characterized in that, The step of determining whether a virtual scheduling partition is an input partition or an output partition based on the power supply and demand balance conditions and the end-of-day water level constraints of the hydropower station for each virtual scheduling partition includes: If the total real-time output of a virtual dispatching zone during the dispatching period is greater than or equal to its load demand, and the water level at the end of the dispatching period of each hydropower station in the zone is not lower than its preset end-of-day water level boundary condition, then the zone is determined to be an output-type zone. If the total real-time output of a virtual dispatching zone during the dispatching period is less than its load demand, or if the water level at the end of the dispatching period of any hydropower station in the zone is lower than its preset end-of-day water level boundary condition in order to meet the load demand, then the zone is determined to be an input-type zone.

6. The method according to claim 5, characterized in that, The objective function of the preset short-term optimization scheduling model is configured as follows: Maximize the net difference between the total electricity sales revenue of all output-type zones and the total electricity purchase cost of all input-type zones during the dispatch period; The pre-defined short-term optimization scheduling model's interval power balance constraint is configured such that the sum of the net output of all output-type zones during any scheduling period is equal to the sum of the net load demand of all input-type zones during the same period. The operational constraints of the hydropower stations within the specified zone are configured as follows: water balance constraints, power generation flow constraints, discharge flow constraints, reservoir capacity constraints, hydraulic connection constraints, power output constraints, water level fluctuation constraints, and initial and final water level constraints.

7. The method according to claim 6, characterized in that, The step of using the virtual scheduling partition as the scheduling unit and employing a preset optimization algorithm to solve a preset short-term optimization scheduling model to obtain the short-term scheduling plan for the cross-basin hydropower station group includes: The short-term optimal scheduling model is solved using a stepwise optimization algorithm. The multi-stage decision problem for the entire scheduling period is decomposed into a series of sequentially executed two-stage optimization sub-problems. Each two-stage sub-problem is optimized and solved, and the short-term scheduling plan is obtained through iterative calculation.

8. A short-term optimized scheduling device for a cross-basin hydropower station group considering virtual scheduling partitions, characterized in that, include: The partitioning module is used to divide the hydropower station group into multiple virtual scheduling zones based on the hydraulic and electrical connection strengths between the hydropower stations in the cross-basin hydropower station group. The discrimination module is used to classify each virtual scheduling partition as an input partition or an output partition based on the power supply and demand balance conditions and the end-of-day water level constraints of the hydropower station. The input partition indicates a power shortage that requires power supply, while the output partition indicates a power surplus that can be transmitted to other regions. The solution module is used to solve the preset short-term optimization scheduling model using the virtual scheduling partition as the scheduling unit and a preset optimization algorithm to obtain the short-term scheduling plan of the cross-basin hydropower station group; wherein the short-term optimization scheduling model is configured to take the maximization of the total system benefit as the objective function, and includes at least the inter-regional power balance constraints and the hydropower station operation constraints within the partition.

9. An electronic device, characterized in that, include: A memory, and one or more processors communicatively connected to the memory; The memory stores instructions that can be executed by the one or more processors to cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images