Hydropower station heterogeneous resource elastic scheduling system based on dynamic priority

Abstract

The invention relates to the technical field of resource scheduling, in particular to a hydropower station heterogeneous resource elastic scheduling system based on dynamic priority, which comprises a data acquisition module for acquiring real-time data corresponding to power resources; a calculation triggering unit in the dynamic priority calculation module constructs a resource state change triggering condition and a data state change triggering condition according to the state of the power resource; a monomer priority calculation unit constructs a monomer priority according to the power resource priority evaluation matrix; the coupling priority calculation unit receives the priority feature vector through a coupling priority prediction model and outputs a coupling priority; the hydropower station constraint priority calculation module calculates the constraint condition priority related to the internal operation of the cascade hydropower station according to the real-time operation state and the internal operation constraint set of the current cascade hydropower station; a scheduling triggering unit in the resource scheduling module is used for triggering resource scheduling; and the scheduling execution unit performs scheduling and optimal distribution of different power resources.

Description

Technical Field

[0001] The present invention relates to the technical field of resource scheduling, and in particular to a hydropower station heterogeneous resource elastic scheduling system based on dynamic priorities. Background Art

[0002] Currently, hydropower station dispatching systems face complex multi-objective optimization requirements and resource heterogeneity challenges. With the diversification of energy structures, hydropower stations often need to operate in coordination with other energy sources such as thermal power and renewable energy to achieve grid stability and economic goals. However, traditional dispatching methods often adopt static priority strategies in the resource allocation process, which makes it difficult to adapt to dynamically changing dispatching needs in different time periods and scenarios. For example, during peak load periods, power supply must be prioritized, while during ecologically sensitive periods, downstream flow control must be taken into account. Existing technologies lack flexible adjustment mechanisms, resulting in limited resource utilization efficiency and difficulty in balancing conflicts among multiple objectives. In addition, the coupling relationship between heterogeneous resources is complex, and a single optimization model cannot effectively integrate hydraulic, power, and environmental constraints, which can easily lead to local optimality or low computational efficiency.

[0003] Therefore, a dynamic priority-based elastic scheduling system for heterogeneous resources of hydropower stations is proposed. Summary of the Invention

[0004] The purpose of the present invention is to provide a hydropower station heterogeneous resource elastic scheduling system based on dynamic priority to improve the efficiency of resource scheduling of hydropower stations. Real-time data corresponding to power resources are collected through a data acquisition module; a calculation trigger unit in a dynamic priority calculation module constructs resource state change trigger conditions and data state change trigger conditions according to the state of power resources; a single priority calculation unit constructs single priority according to a power resource priority evaluation matrix; a coupling priority calculation unit receives a priority feature vector through a coupling priority prediction model and outputs a coupling priority; a hydropower station constraint priority calculation module calculates the priority of constraint conditions related to the internal operation of a cascade hydropower station according to the real-time operating state and internal operating constraint set of the current cascade hydropower station; a scheduling trigger unit in a resource scheduling module is used to trigger resource scheduling; and a scheduling execution unit performs scheduling and optimal allocation of different power resources.

[0005] To achieve the above object, the present invention provides the following technical solutions: A hydropower station heterogeneous resource elastic scheduling system based on dynamic priority, comprising: Data acquisition module, used to collect real-time data corresponding to power resources; Furthermore, the data acquisition module includes: Hydropower resources, wind power resources, photovoltaic resources and thermal power resources constitute a heterogeneous resource set; Collect real-time data on each power resource in a heterogeneous resource set, including hydrological, meteorological, grid load, power market prices, equipment operating status, constraints, and ecological and environmental monitoring data; The collected data is cleaned, verified and integrated to obtain the real-time resource state vector of each power resource.

[0006] The dynamic priority calculation module includes a calculation trigger unit, a single-unit priority calculation unit, and a coupling priority calculation unit. The calculation trigger unit constructs resource state change trigger conditions and data state change trigger conditions based on the state of the power resource; the single-unit priority calculation unit constructs single-unit priorities based on the power resource priority evaluation matrix; the coupling priority calculation unit receives the priority feature vector through the coupling priority prediction model and outputs the coupling priority; Furthermore, the calculation trigger unit includes: Resource status change trigger conditions and data status change trigger conditions, where resource status change trigger conditions include triggering calculations when power resources reach a preset state or cross a warning line, and triggering calculations when external events occur; If the resource status change trigger conditions are not met, the priority calculation trigger is performed according to the data status change trigger conditions; the data status change trigger conditions include forecast update trigger, real-time data anomaly trigger and market price fluctuation trigger, among which the forecast update trigger includes triggering calculation when the forecast deviation between the current forecast data of the power resource and the corresponding data in the historical period exceeds the threshold; Real-time data anomaly triggering involves building a key parameter prediction model. By analyzing the relationship between input parameters and output indicators, the parameters most sensitive to the output indicators are identified, and the key parameters of the power resources are obtained. The change rate of the key parameters is monitored, and if the change rate threshold is exceeded, the priority calculation is triggered. The market price fluctuation trigger includes the calculation of trigger priority when the real-time or predicted electricity market price fluctuation exceeds a set percentage or absolute value threshold.

[0007] Furthermore, the monomer priority calculation unit includes: The real-time resource state vector of power resources is subjected to resource availability analysis, load demand analysis, market environment analysis, constraint state analysis and environmental ecological state analysis to construct a power resource priority evaluation matrix; the power resource priority evaluation matrix is transposed and multiplied to construct a semi-positive definite power resource priority evaluation matrix, and the sum of the eigenvalues of the matrix is calculated and marked as the real-time individual priority of the power resource.

[0008] Furthermore, the coupling priority calculation unit includes: Obtain historical data for each power resource, including the output, reservoir water level, inflow, and outflow of each cascade hydropower station; the output and wind speed of each wind farm; the output and sunlight intensity of each photovoltaic power station; the output and start / stop status of each thermal power plant; and the total grid load and market electricity price. Clean the collected historical data and process missing values and outliers; Features that reflect the state and interrelationships of heterogeneous resource combinations are extracted from historical data. These features include resource output ratios, resource output change rates, resource coupling characteristics, and system state characteristics, and a priority feature vector is constructed. Resource coupling characteristics reflect the coordination or conflict between resources, including covariance, response amplitude, and the degree of hydropower participation in peak regulation, frequency regulation, or backup services. System state characteristics reflect the characteristics of the macro-system state, including load levels, market price levels, reservoir storage rates, seasons, and time periods. The coupling priority prediction model receives the priority feature vector, outputs the coupling priority of water-wind, water-light, and water-fire resources, and sorts them.

[0009] The hydropower station constraint priority calculation module calculates the priority of the constraints related to the internal operation of the cascade hydropower station based on the real-time operating status and internal operating constraint set of the current cascade hydropower station; Furthermore, the hydropower station constraint priority calculation module includes: calculating the priority of the constraint conditions related to the internal operation of the cascade hydropower station based on the real-time operating status and internal operating constraint set of the current cascade hydropower station, and the internal operating constraint set includes at least reservoir safety constraints, ecological and environmental protection constraints, unit operation constraints, hydraulic connection constraints, and hydraulic structure constraints.

[0010] The resource scheduling module includes a scheduling trigger unit and a scheduling execution unit, wherein the scheduling trigger unit is used to trigger resource scheduling; the scheduling execution unit first performs the first resource allocation according to the coupling priority, then performs power resource scheduling according to the single-unit priority, and finally performs resource scheduling and optimal allocation within the cascade hydropower station.

[0011] Further, the scheduling execution unit receives the sorted coupling priorities and performs the first resource allocation according to the coupling priorities; The scheduling execution unit obtains the monomer priority of the corresponding power resource according to the result of the first resource allocation, and schedules hydropower and other heterogeneous resources according to the monomer priority; The dispatch execution unit performs resource dispatch and optimal allocation within the cascade hydropower station according to the constraint priority of the hydropower station constraint priority calculation module and the received dispatch instructions, and generates a cascade operation plan; the internal dispatch and allocation process includes but is not limited to: cascade joint optimization, unit combination and load distribution, reservoir operation fine management, and auxiliary service response allocation.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. The calculation trigger unit achieves event-driven activation of dynamic priority calculations by accurately defining two types of trigger conditions: resource status changes and data status changes. Calculations are started only when they are truly needed, effectively avoiding the high-frequency computing overhead and resource waste caused by continuous monitoring, thereby significantly reducing operating costs and improving overall system efficiency.

[0013] 2. Single-unit priority calculation can quickly respond to changes in load and resource status at the micro level by conducting refined evaluations of multi-dimensional indicators such as real-time resource availability, load demand, market environment, constraint status, and ecological status. Coupled priority calculation uses output ratio, change rate, coordination / conflict characteristics, and system macro-state characteristics to construct feature vectors, which can capture the interactions between multiple resources such as water-wind, water-light, and water-fire, and achieve overall collaborative optimization priority evaluation.

[0014] 3. By allocating resources based on coupling priority and single-unit priority in turn through the dispatching execution unit and combining it with the dynamic constraint priority within the cascade station, the system can achieve rapid response and refined resource allocation for key power generation, peak regulation and auxiliary service tasks, significantly improving the flexibility and real-time performance of task execution. In the complex environment of multi-time domain and multi-physical field coupling, this mechanism effectively reduces the phenomenon of water or wind and solar power abandonment caused by fluctuations in renewable energy through refined unit combination and load distribution, significantly reduces load loss, and improves the robustness of grid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the structure of a hydropower station heterogeneous resource elastic scheduling system based on dynamic priority provided by an embodiment of the present invention; Figure 2 A flowchart of calculating monomer priority provided by an embodiment of the present invention; Figure 3 A flowchart of calculating coupling priority provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention. Example 1:

[0017] In order to realize the flexible dispatch of various heterogeneous power resources such as hydropower, thermal power, wind power and photovoltaic power generation, and improve the efficiency of resource dispatch of hydropower stations, a power company introduced a hydropower station heterogeneous resource flexible dispatch system based on dynamic priority provided by the present invention. The system structure is as follows: Figure 1 As shown, the specific implementation is as follows: Data acquisition module, used to collect real-time data corresponding to power resources; Furthermore, the data acquisition module includes: Hydropower resources, wind power resources, photovoltaic resources and thermal power resources constitute a heterogeneous resource set; Collect real-time data on each power resource in a heterogeneous resource set, including hydrological, meteorological, grid load, power market prices, equipment operating status, constraints, and ecological and environmental monitoring data; The collected data is cleaned, verified and integrated to obtain the real-time resource state vector of each power resource.

[0018] Furthermore, the heterogeneous resource collection includes at least hydropower resources composed of cascade hydropower stations, wind power resources composed of wind farms, and photovoltaic resources composed of solar photovoltaic power stations, and the collection can also selectively include thermal power resources composed of thermal power plants, as well as other resource types with scheduling potential (such as energy storage systems, controllable loads, etc.), forming a joint scheduling entity for multiple types of energy resources.

[0019] Furthermore, data collection is achieved through various technical means such as sensors, telemetry systems, communication networks, and data interfaces. The collected raw real-time data is cleaned, verified, and preprocessed automatically or manually, including data cleaning: identifying and processing missing values, outliers, noise, or erroneous data in the data; data verification: checking the logical consistency of the data, for example, checking the balance of water inflow and outflow in the reservoir, checking whether the unit output is within its operating range, etc.; data integration: unifying data from different data sources, different sampling frequencies, and different formats, and aligning timestamps to form a structured, standardized real-time data set.

[0020] Furthermore, based on the cleaned, verified, and integrated data, a real-time resource state vector is generated that accurately reflects the current operational status of each heterogeneous resource for use by subsequent modules. Table 1 shows some data for the real-time resource state vector for hydropower resources.

[0021] Table 1. Real-time resource state vector of hydropower resources

[0022] By building a comprehensive, real-time data collection system that covers multiple types of power resources and integrates multiple data dimensions, it is possible to capture real-time changes in the operating status of heterogeneous resources, the external environment, and grid demand, providing a solid data foundation for subsequent dynamic priority calculations and flexible scheduling, enabling the system to make scheduling decisions based on the most timely and accurate information.

[0023] The dynamic priority calculation module includes a calculation trigger unit, a single-unit priority calculation unit, and a coupling priority calculation unit. The calculation trigger unit constructs resource state change trigger conditions and data state change trigger conditions based on the state of the power resource; the single-unit priority calculation unit constructs single-unit priorities based on the power resource priority evaluation matrix; the coupling priority calculation unit receives the priority feature vector through the coupling priority prediction model and outputs the coupling priority; Furthermore, the calculation trigger unit includes: Resource state change trigger conditions and data state change trigger conditions, where resource state change trigger conditions include triggering calculations when power resources reach a specific state or cross a warning line, and triggering calculations when external events occur; If the resource status change trigger conditions are not met, the priority calculation trigger is performed according to the data status change trigger conditions; the data status change trigger conditions include forecast update trigger, real-time data anomaly trigger and market price fluctuation trigger, among which the forecast update trigger includes triggering calculation when the forecast deviation between the current forecast data of the power resource and the corresponding data in the historical period exceeds the threshold; Real-time data anomaly triggering involves building a key parameter prediction model. By analyzing the relationship between input parameters and output indicators, the parameters most sensitive to the output indicators are identified, and the key parameters of the power resources are obtained. The change rate of the key parameters is monitored, and if the change rate threshold is exceeded, the priority calculation is triggered. The market price fluctuation trigger includes the calculation of trigger priority when the real-time or predicted electricity market price fluctuation exceeds a set percentage or absolute value threshold.

[0024] Furthermore, the forecast update trigger includes triggering calculation when an update of any key forecast data (including but not limited to hydrological forecast, meteorological forecast, power grid load forecast, electricity market price forecast) is received, and the forecast value deviation (absolute deviation or relative deviation) of the future key time period (such as the next 1 hour, the next 6 hours or the next 24 hours) compared with the previous round of forecast data exceeds a set threshold, where the forecast data is data obtained through existing forecasting means.

[0025] Furthermore, a key parameter prediction model is constructed to analyze the relationship between various input parameters in the system operation data (such as real-time inflow, real-time wind speed, real-time light intensity, and real-time load) and key output indicators (such as system load). By analyzing the impact of each input parameter in the key parameter prediction model on the output indicator (for example, constructing a key parameter prediction model through sensitivity analysis, feature importance assessment, etc.), the "key parameters of power resources" that have the most significant impact on system operation or scheduling results are identified. The actual value or rate of change of these key parameters is monitored in real time. If the rate of change (for example, the ratio of the value change per unit time to its mean or threshold) exceeds a set dynamic or static rate of change threshold, the calculation of dynamic priority is triggered.

[0026] The computation trigger unit establishes an intelligent, multi-level trigger mechanism that can accurately determine when the priority needs to be recalculated, thereby avoiding unnecessary repeated calculations and improving the system's computing efficiency. At the same time, it can promptly respond to sudden changes in the system's operating status, abnormal fluctuations in key data, and the impact of external events, ensuring that dynamic priorities always reflect current actual needs and potential risks, providing timely and accurate decision-making basis for elastic scheduling, and greatly enhancing the real-time, robustness, and adaptability of the entire heterogeneous resource elastic scheduling system to complex dynamic environments.

[0027] Furthermore, the calculation process of the monomer priority is as follows Figure 2 As shown, the monomer priority calculation unit includes: The real-time resource state vector of power resources is subjected to resource availability analysis, load demand analysis, market environment analysis, constraint state analysis and environmental ecological state analysis to construct a power resource priority evaluation matrix; the power resource priority evaluation matrix is transposed and multiplied to construct a semi-positive definite power resource priority evaluation matrix, and the sum of the eigenvalues of the matrix is calculated and marked as the real-time individual priority of the power resource.

[0028] Furthermore, a multi-dimensional analysis is performed on the received real-time resource state vector for each heterogeneous power resource. This analysis includes at least an assessment of the power resource's resource availability (e.g., adjustable water volume and output capacity for hydropower, real-time output and forecast uncertainty for wind / photovoltaic power, and available capacity and start / stop status for thermal power), current grid load demand (load-bearing potential associated with the resource type), the power market environment (e.g., the real-time revenue or cost of generating electricity from the resource), its constraints (e.g., water level limits for hydropower stations, unit output limits, ramp rates for thermal power units), and its environmental and ecological status (e.g., the degree to which ecological flow requirements are met). Based on the results of this multi-dimensional analysis, a power resource priority evaluation matrix is constructed, reflecting the current comprehensive priority of the power resource. This evaluation matrix quantifies the performance or importance of the power resource across different evaluation dimensions.

[0029] Furthermore, the power resource priority evaluation matrix and its transposed matrix are multiplied to construct a semi-positive definite power resource priority evaluation matrix; then, the eigenvalues of the semi-positive definite power resource priority evaluation matrix are calculated, and the sum of the eigenvalues (or other mathematical indicators that can comprehensively reflect the overall "importance" or "energy" of the matrix, such as the maximum eigenvalue, etc.) is obtained, and the sum of the eigenvalues is marked as the real-time individual priority value of the power resource at the current moment.

[0030] Based on comprehensive real-time status data, through structured multi-dimensional analysis and rigorous matrix calculation methods, an objective, quantitative individual priority value is calculated for each heterogeneous power resource. This quantitative assessment takes into account the resource's own operating characteristics, current availability, and interaction with the external environment, providing important basic information for subsequent hierarchical scheduling.

[0031] Furthermore, the calculation process of coupling priority is as follows Figure 3 As shown, the coupling priority calculation unit includes: Obtain historical data for each power resource, including the output, reservoir water level, inflow, and outflow of each cascade hydropower station; the output and wind speed of each wind farm; the output and sunlight intensity of each photovoltaic power station; the output and start / stop status of each thermal power plant; and the total grid load and market electricity price. Clean the collected historical data and process missing values and outliers; Features that reflect the state and interrelationships of heterogeneous resource combinations are extracted from historical data. These features include resource output ratios, resource output change rates, resource coupling characteristics, and system state characteristics, and a priority feature vector is constructed. Resource coupling characteristics reflect the coordination or conflict between resources, including covariance, response amplitude, and the degree of hydropower participation in peak regulation, frequency regulation, or backup services. System state characteristics reflect the characteristics of the macro-system state, including load levels, market price levels, reservoir storage rates, seasons, and time periods. The coupling priority prediction model receives the priority feature vector, outputs the coupling priority of water-wind, water-light, and water-fire resources, and sorts them.

[0032] Furthermore, the coupling priority prediction model receives the constructed priority feature vector as input and, through predictive logic or algorithms trained or constructed based on historical data (e.g., machine learning models such as regression models or neural networks, or complex statistical analysis or rule-based reasoning), outputs numerical priorities reflecting the importance of synergy or conflict between different heterogeneous resource combinations (e.g., water-wind coupling, water-solar coupling, water-fire coupling, etc.) at the current moment or in the short future. The coupling priority values of each heterogeneous resource combination output by the prediction model are then ranked to determine the heterogeneous resource coupling relationships that require the most priority attention and coordination in the current state. Table 2 shows the coupling priority ranking results at a specific moment.

[0033] Table 2. Coupling priority ranking results

[0034] By constructing and utilizing a coupling priority prediction model, the system can dynamically and quantitatively evaluate the importance and coordination needs of different resource combinations in the current environment based on the real-time changing status of heterogeneous resources; this dynamic coupling priority ranking result is used to guide subsequent macro-resource scheduling, enabling the system to more intelligently balance the relationship between different types of resources. For example, when wind and solar power fluctuate greatly, it can focus on coordinating hydropower and wind power, and during peak load periods, it can focus on coordinating water and fire output, thereby improving the scientific nature and pertinence of macro-scheduling decisions and laying the foundation for achieving overall coordinated optimization of heterogeneous resources.

[0035] The hydropower station constraint priority calculation module calculates the priority of the constraints related to the internal operation of the cascade hydropower station based on the real-time operating status and internal operating constraint set of the current cascade hydropower station; Furthermore, the hydropower station constraint priority calculation module includes: calculating the priority of the constraint conditions related to the internal operation of the cascade hydropower station based on the real-time operating status and internal operating constraint set of the current cascade hydropower station, and the internal operating constraint set includes at least reservoir safety constraints, ecological and environmental protection constraints, unit operation constraints, hydraulic connection constraints, and hydraulic structure constraints.

[0036] Furthermore, the core function of the hydropower station constraint priority calculation module is to dynamically calculate the priority levels of various constraint conditions closely related to the internal operation of the cascade hydropower station based on the real-time operating status and detailed internal constraint set of the current cascade hydropower station, and to carry out refined scheduling and resource allocation within the cascade hydropower station based on these priorities.

[0037] Furthermore, the current real-time operating status of the cascade hydropower station group includes the real-time water level of each reservoir, inflow flow, executed outflow flow, real-time output, operating conditions and health status of each unit, as well as power consumption within the plant.

[0038] Furthermore, the priority of constraints changes dynamically. For example, when a major flood is forecast and reservoir water levels rise rapidly, approaching the flood control limit, the priority of "reservoir safety constraints" (particularly the flood control limit) will instantly rise to the highest priority, potentially surpassing all other power generation or ecological objectives. During extreme dry seasons or when required by law, the priority of "ecological and environmental protection constraints" (particularly the minimum ecological flow) will be significantly increased. When a critical unit fails or undergoes emergency maintenance, ensuring the safe operation of other units and stable power supply to the system will take higher priority. This priority calculation process can be implemented based on rule-based reasoning, fuzzy logic, risk assessment models, or weighting functions related to real-time conditions.

[0039] This module ensures the safe, efficient, and flexible operation of cascade hydropower stations within their complex constraints. It prioritizes the most critical internal operational requirements, particularly when responding to extreme water conditions (such as floods and low water levels), ensuring ecological flows, and meeting the demands of refined grid regulation. This prioritized, refined internal management maximizes the role of cascade hydropower stations as core regulating power sources, improving their adaptability to uncertainty and overall operational efficiency. It is essential for the reliable operation of a heterogeneous resource elastic scheduling system.

[0040] The resource scheduling module includes a scheduling trigger unit and a scheduling execution unit, wherein the scheduling trigger unit is used to trigger resource scheduling; the scheduling execution unit first performs the first resource allocation according to the coupling priority, then performs power resource scheduling according to the single-unit priority, and finally performs resource scheduling and optimal allocation within the cascade hydropower station.

[0041] Furthermore, the scheduling trigger unit works in conjunction with the calculation trigger unit of the dynamic priority calculation module to ensure that the scheduling execution unit is started in time when the dynamic priority changes significantly, the system status becomes abnormal, or the preset scheduling cycle is reached.

[0042] Furthermore, the scheduling execution unit receives the sorted coupling priorities and executes a first resource allocation based on the coupling priorities. The system receives the sorted coupling priority information output by the coupling priority calculation unit and executes the first resource allocation based on the highest-priority heterogeneous resource coupling relationship, thereby making a macro-level resource scheduling decision. This first resource allocation determines the overall strategy and preliminary output ranges for different types of heterogeneous resources (hydro, wind, solar, thermal, etc.) in terms of total load sharing, coordinated peak / frequency regulation, backup provision, and overall energy balance, guiding the overall direction of coordination among these different resource types. For example, if hydro-wind coupling has the highest priority, macro-level consideration will be given to how to leverage hydropower to complement the volatility of wind power.

[0043] Furthermore, the scheduling execution unit obtains the individual priorities of the corresponding power resources based on the result of the first resource allocation, and schedules hydropower and other heterogeneous resources based on the individual priorities; and obtains the individual priority information of each corresponding power resource based on the result of the first resource allocation. Under the overall framework determined by the macro allocation, medium-granularity power resource scheduling optimization is performed based on the individual priorities of each power resource. The process aims to determine the specific output plan and operating status of a single hydropower station, wind farm, photovoltaic power station, thermal power plant, etc. in accordance with the requirements of the macro strategy, taking into full consideration the inherent characteristics, operating limitations and individual priorities of each individual resource. For example, when the macro strategy requires an overall increase in hydropower generation, hydropower stations with high individual priorities will be given priority in increasing their output.

[0044] Furthermore, the dispatching execution unit performs resource dispatching and optimal allocation within the cascade hydropower station according to the constraint priority of the hydropower station constraint priority calculation module and the received dispatching instructions, and generates a cascade operation plan; the internal dispatching and allocation process includes but is not limited to: cascade joint optimization, unit combination and load distribution, fine management of reservoir operation, and auxiliary service response distribution.

[0045] Furthermore, the system receives the constraint priority output from the hydropower station constraint priority calculation module, as well as the overall dispatching instructions (such as total cascade output and total downstream flow requirements) issued by the upper-level dispatcher (including the first resource allocation results and the dispatching results based on individual priorities) for the cascade hydropower station. Based on this information, the system performs the most detailed and refined resource scheduling and optimization allocation within the cascade hydropower station, generating a final executable cascade operation plan. The internal scheduling and allocation process includes, but is not limited to: cascade joint optimization solution based on internal constraint priorities (optimizing the water level, flow, and output process of each power station while satisfying upper-level instructions and high-priority internal constraints), unit combination and load fine-tuning (determining the start-up units and specific output of each power station), fine-tuning reservoir operation management (precisely controlling reservoir water level and downstream flow to meet multi-objective requirements), and internal task allocation for auxiliary service responses.

[0046] The resource scheduling module builds an intelligent execution system that can perform hierarchical scheduling based on priority information. By decomposing complex heterogeneous resource scheduling problems into three levels: macro, single-unit, and internal, and using different types of dynamic priorities (coupling priority, single-unit priority, internal constraint priority) to guide optimization decisions at each level, the system can effectively handle large-scale, multi-objective, and strongly constrained scheduling problems.

[0047] The present invention overcomes the shortcomings of traditional scheduling methods in dealing with large-scale, multi-objective, and strongly constrained joint scheduling of heterogeneous resources by constructing a system that integrates multi-source data collection, dynamic priority intelligent calculation, and hierarchical elastic scheduling. In particular, the calculation of single-unit priority and coupling priority provides the system with the ability to understand and weigh resource characteristics at different levels, while the hydropower station constraint priority calculation module and internal scheduling allocation ensure the internal operation safety and efficiency of cascade hydropower stations as core regulating resources. Ultimately, the hierarchical decoupling and priority-driven execution of the resource scheduling module improve the efficiency and convergence of scheduling, so that the generated scheduling plan can better balance the relationship between different resource types, flexibly respond to uncertainty, and at the same time ensure the safety and precision of the internal operation of cascade hydropower stations, thereby significantly improving the overall performance of the entire heterogeneous resource elastic scheduling system. Example 2:

[0048] To achieve flexible scheduling of heterogeneous power resources, a power company introduced a hydropower station heterogeneous resource flexible scheduling system based on dynamic priority provided by the present invention. The specific implementation method is as follows: Data acquisition module, used to collect real-time data corresponding to power resources; Furthermore, the data acquisition module includes: Hydropower resources, wind power resources, photovoltaic resources and thermal power resources constitute a heterogeneous resource set; Collect real-time data on each power resource in a heterogeneous resource set, including hydrological, meteorological, grid load, power market prices, equipment operating status, constraints, and ecological and environmental monitoring data; The collected data is cleaned, verified and integrated to obtain the real-time resource state vector of each power resource.

[0049] The dynamic priority calculation module includes a calculation trigger unit, a single-unit priority calculation unit, and a coupling priority calculation unit. The calculation trigger unit constructs resource state change trigger conditions and data state change trigger conditions based on the state of the power resource; the single-unit priority calculation unit constructs single-unit priorities based on the power resource priority evaluation matrix; the coupling priority calculation unit receives the priority feature vector through the coupling priority prediction model and outputs the coupling priority; Furthermore, the calculation trigger unit includes: Resource status change trigger conditions and data status change trigger conditions, where resource status change trigger conditions include triggering calculations when power resources reach a preset state or cross a warning line, and triggering calculations when external events occur; If the resource status change trigger conditions are not met, the priority calculation trigger is performed according to the data status change trigger conditions; the data status change trigger conditions include forecast update trigger, real-time data anomaly trigger and market price fluctuation trigger, among which the forecast update trigger includes triggering calculation when the forecast deviation between the current forecast data of the power resource and the corresponding data in the historical period exceeds the threshold; Real-time data anomaly triggering involves building a key parameter prediction model. By analyzing the relationship between input parameters and output indicators, the parameters most sensitive to the output indicators are identified, and the key parameters of the power resources are obtained. The change rate of the key parameters is monitored, and if the change rate threshold is exceeded, the priority calculation is triggered. The market price fluctuation trigger includes the calculation of trigger priority when the real-time or predicted electricity market price fluctuation exceeds a set percentage or absolute value threshold.

[0050] Furthermore, the monomer priority calculation unit includes: The real-time resource state vector of power resources is subjected to resource availability analysis, load demand analysis, market environment analysis, constraint state analysis and environmental ecological state analysis to construct a power resource priority evaluation matrix; the power resource priority evaluation matrix is transposed and multiplied to construct a semi-positive definite power resource priority evaluation matrix, and the sum of the eigenvalues of the matrix is calculated and marked as the real-time individual priority of the power resource.

[0051] Furthermore, the coupling priority calculation unit includes: Obtain historical data for each power resource, including the output, reservoir water level, inflow, and outflow of each cascade hydropower station; the output and wind speed of each wind farm; the output and sunlight intensity of each photovoltaic power station; the output and start / stop status of each thermal power plant; and the total grid load and market electricity price. Clean the collected historical data and process missing values and outliers; Features that reflect the state and interrelationships of heterogeneous resource combinations are extracted from historical data. These features include resource output ratios, resource output change rates, resource coupling characteristics, and system state characteristics, and a priority feature vector is constructed. Resource coupling characteristics reflect the coordination or conflict between resources, including covariance, response amplitude, and the degree of hydropower participation in peak regulation, frequency regulation, or backup services. System state characteristics reflect the characteristics of the macro-system state, including load levels, market price levels, reservoir storage rates, seasons, and time periods. The coupling priority prediction model receives the priority feature vector, outputs the coupling priority of water-wind, water-light, and water-fire resources, and sorts them.

[0052] The hydropower station constraint priority calculation module calculates the priority of the constraints related to the internal operation of the cascade hydropower station based on the real-time operating status and internal operating constraint set of the current cascade hydropower station; Furthermore, the hydropower station constraint priority calculation module includes: calculating the priority of the constraint conditions related to the internal operation of the cascade hydropower station based on the real-time operating status and internal operating constraint set of the current cascade hydropower station, and the internal operating constraint set includes at least reservoir safety constraints, ecological and environmental protection constraints, unit operation constraints, hydraulic connection constraints, and hydraulic structure constraints.

[0053] Furthermore, the internal operation constraint set includes at least: Reservoir safety constraints: These include the maximum flood control limit water level, minimum dead water level limit, flood control storage capacity requirements of each reservoir, and the upper and lower limits of the safe operating water level range derived from a dynamic assessment of real-time water conditions and safety monitoring data. These are constraints closely related to the safe operation of dams and related hydraulic structures.

[0054] Ecological and environmental protection constraints: such as mandatory requirements for minimum ecological flow in downstream river sections, requirements for flow processes or water level change rates during specific ecologically sensitive periods (such as fish breeding and migration periods), restrictions on the impact of reservoir discharge on downstream water temperature and water quality, and other operational restrictions related to the protection of aquatic life and the ecological health of the river.

[0055] Unit operation constraints: such as the minimum technical output of a single generator unit or pump station unit in each hydropower station, maximum output limit, safe start / shutdown number limit, minimum continuous operation time, minimum continuous shutdown time, ramp rate limit, and available capacity constraints caused by equipment status monitoring or maintenance plans.

[0056] Hydraulic connection constraints: a strict water balance relationship between cascade hydropower stations, that is, the total downstream flow (the sum of the power generation flow and the water discharge flow) of the upstream power station in each period is equal to or determines the inflow flow of the downstream adjacent power station (excluding the interval water inflow), as well as the functional or tabular relationship between reservoir water level and storage capacity, water level and inflow and outflow, and hydraulic head and output.

[0057] Hydraulic structure constraints: such as the maximum flow capacity and minimum opening requirements of hydraulic structures such as spillways, bottom outlets, diversion tunnels, and tailraces, as well as other operational restrictions related to the safe operation of these structures.

[0058] The resource scheduling module includes a scheduling trigger unit and a scheduling execution unit, wherein the scheduling trigger unit is used to trigger resource scheduling; the scheduling execution unit first performs the first resource allocation according to the coupling priority, then performs power resource scheduling according to the single-unit priority, and finally performs resource scheduling and optimal allocation within the cascade hydropower station.

[0059] Further, the scheduling execution unit receives the sorted coupling priorities and performs the first resource allocation according to the coupling priorities; The scheduling execution unit obtains the monomer priority of the corresponding power resource according to the result of the first resource allocation, and schedules hydropower and other heterogeneous resources according to the monomer priority; The dispatch execution unit dispatches and optimizes resources within the cascade hydropower station based on the constraint priority of the hydropower station constraint priority calculation module and the dispatch instructions it receives, generating a cascade operation plan. This internal dispatch and allocation process includes, but is not limited to, cascade joint optimization, unit commitment and load allocation, refined reservoir operation management, and ancillary service response allocation. Table 3 shows the results of the internal dispatch and optimization allocation of hydropower stations based on their constraint priorities.

[0060] Table 3. Internal scheduling results

[0061] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A hydropower station heterogeneous resource elastic scheduling system based on dynamic priority, characterized by: include: Data acquisition module, used to collect real-time data corresponding to power resources; The dynamic priority calculation module includes a calculation trigger unit, a single priority calculation unit and a coupling priority calculation unit, wherein the calculation trigger unit constructs a resource state change trigger condition and a data state change trigger condition according to the state of the power resource, and the calculation trigger unit includes: a resource state change trigger condition and a data state change trigger condition, wherein the resource state change trigger condition includes triggering calculation when the power resource reaches a preset state or crosses the warning line, and triggering calculation when an external event occurs; if the resource state change trigger condition is not met, the priority calculation trigger is performed according to the data state change trigger condition; the data state change trigger condition includes a forecast update trigger, a real-time data anomaly trigger and a market price fluctuation trigger, wherein the forecast update trigger includes a power resource trigger. The calculation is triggered when the prediction deviation between the current prediction data and the corresponding data in the historical period exceeds the threshold; the real-time data abnormality trigger includes building a key parameter prediction model, identifying the input parameters most sensitive to the output indicators by analyzing the relationship between each input parameter and the output indicator, and obtaining the key parameters of the power resources; monitoring the change rate of the key parameters, if it exceeds the change rate threshold, triggering the priority calculation; market price fluctuation trigger includes triggering the priority calculation when the real-time or predicted power market price fluctuation exceeds the set percentage or absolute value threshold; the single priority calculation unit constructs the single priority according to the power resource priority evaluation matrix; the coupling priority calculation unit receives the priority feature vector through the coupling priority prediction model and outputs the coupling priority; The hydropower station constraint priority calculation module calculates the priority of the constraints related to the internal operation of the cascade hydropower station based on the real-time operating status and internal operating constraint set of the current cascade hydropower station; The resource scheduling module includes a scheduling trigger unit and a scheduling execution unit, wherein the scheduling trigger unit is used to trigger resource scheduling; the scheduling execution unit first performs the first resource allocation according to the coupling priority, then performs power resource scheduling according to the single-unit priority, and finally performs resource scheduling and optimal allocation within the cascade hydropower station.

2. A hydropower station heterogeneous resource elastic scheduling system based on dynamic priority according to claim 1, characterized in that: The data acquisition module includes: Hydropower resources, wind power resources, photovoltaic resources and thermal power resources constitute a heterogeneous resource set; Collect real-time data on each power resource in a heterogeneous resource set, including hydrological, meteorological, grid load, power market prices, equipment operating status, constraints, and ecological and environmental monitoring data; The collected data is cleaned, verified and integrated to obtain the real-time resource state vector of each power resource.

3. The hydropower station heterogeneous resource elastic scheduling system based on dynamic priority according to claim 1 is characterized in that: The single priority calculation unit includes: The real-time resource state vector of power resources is subjected to resource availability analysis, load demand analysis, market environment analysis, constraint state analysis and environmental ecological state analysis to construct a power resource priority evaluation matrix; the power resource priority evaluation matrix is transposed and multiplied to construct a semi-positive definite power resource priority evaluation matrix, and the sum of the eigenvalues of the semi-positive definite power resource priority evaluation matrix is calculated and marked as the real-time individual priority of the power resource.

4. The hydropower station heterogeneous resource elastic scheduling system based on dynamic priority according to claim 1 is characterized in that: The coupling priority calculation unit includes: Obtain historical data for each power resource, including the output, reservoir water level, inflow, and outflow of each cascade hydropower station; the output and wind speed of each wind farm; the output and sunlight intensity of each photovoltaic power station; the output and start / stop status of each thermal power plant; and the total grid load and market electricity price. Clean the collected historical data and process missing values and outliers; Features that reflect the state and interrelationships of heterogeneous resource combinations are extracted from historical data. These features include resource output ratios, resource output change rates, resource coupling characteristics, and system state characteristics, and a priority feature vector is constructed. Resource coupling characteristics reflect the coordination or conflict between resources, including covariance, response amplitude, and the degree of hydropower participation in peak regulation, frequency regulation, or backup services. System state characteristics reflect the characteristics of the macro-system state, including load levels, market price levels, reservoir storage rates, seasons, and time periods. The coupling priority prediction model receives the priority feature vector, outputs the coupling priority of water-wind, water-light, and water-fire resources, and sorts them.

5. The hydropower station heterogeneous resource elastic scheduling system based on dynamic priority according to claim 1 is characterized in that: The hydropower station constraint priority calculation module includes: calculating the priority of the constraint conditions related to the internal operation of the cascade hydropower station based on the current real-time operating status and internal operating constraint set of the cascade hydropower station. The internal operating constraint set at least includes reservoir safety constraints, ecological and environmental protection constraints, unit operation constraints, hydraulic connection constraints, and hydraulic structure constraints.

6. The hydropower station heterogeneous resource elastic scheduling system based on dynamic priority according to claim 1 is characterized in that: The scheduling execution unit receives the sorted coupling priorities and executes a first resource allocation according to the coupling priorities; The scheduling execution unit obtains the monomer priority of the corresponding power resource according to the result of the first resource allocation, and schedules hydropower and other heterogeneous resources according to the monomer priority; The dispatch execution unit performs resource dispatch and optimization allocation within the cascade hydropower station according to the constraint priority of the hydropower station constraint priority calculation module and the received dispatch instructions, and generates a cascade operation plan; The internal scheduling and allocation process includes but is not limited to: cascade joint optimization, unit combination and load distribution, reservoir operation fine management, and auxiliary service response allocation.

Images