A multi-level power dispatching system-oriented regulation capacity fusion calculation method and system

Abstract

The application discloses a kind of multi-level power dispatching system-oriented regulation capacity fusion calculation method and system, method includes: demand side calculation is according to load, unadjustable power output and power data calculation regulation load, and based on regulation load calculation demand side's index;Supply side calculation is according to power type calculation each type power regulation capacity;The regulation capacity of power type calculated is gradually accumulated as whole network inherent, automatic, manual regulation capacity by hierarchical aggregation;According to the regulation load calculated, the index of demand side and the regulation capacity of each type power, regulation margin is calculated, and regulation margin evaluation result is obtained;By configuration file initialization step demand side calculation, supply side calculation, hierarchical aggregation and the index calculation function mapping and calculation order required by regulation margin calculation, drive each step sequentially executes.The application can realize the automatic, hierarchical, configurable calculation and monitoring of the regulation demand within a day and the regulation capacity of multiple types of power.

Description

Technical Field

[0001] This invention relates to the field of power grid dispatch automation technology, and in particular to a method and system for calculating the regulation capacity of a multi-level power dispatch system. Background Technology

[0002] With the rapid increase in the proportion of renewable energy, the quantitative assessment of the daily regulation demand and capacity of the power system has become crucial to ensuring the safe and stable operation of the power grid. Significant fluctuations in renewable energy sources and supply-demand imbalances can both cause the system frequency to deviate from the baseline value for extended periods. Therefore, it is necessary to monitor system regulation demand in a timely manner and simultaneously assess the regulation capacity of various power sources to support dispatching personnel in making intraday frequency regulation decisions.

[0003] Currently, power dispatch automation systems generally use manually defined formulas for measurement points to perform statistical calculations. While this method is intuitive and easy to understand, it presents several problems as the power grid expands, the types of regulating power sources increase (thermal power, hydropower, pumped storage, energy storage, nuclear power, etc.), and the demand for multi-regional and multi-level dispatch coordination grows: First, manual maintenance is labor-intensive, requiring formula configuration for each new indicator; second, it is difficult to adapt to the differentiated calculation rules of different provincial power grids; third, it lacks an automatic processing mechanism for complex dependencies between indicators; and fourth, it cannot achieve automatic aggregation from equipment to the entire network at each level. Summary of the Invention

[0004] Purpose of the invention: The purpose of this invention is to provide a method and system for calculating the regulation capacity of a multi-level power dispatching system. By performing hierarchical automatic calculation and fusion statistics on the daily regulation demand of the power grid and the regulation capacity of multiple types of power sources, this invention solves the problem of rapid statistics of real-time data and multi-level collaborative monitoring of large-scale power grids.

[0005] Technical solution: The method for fusion calculation of regulation capacity for multi-level power dispatching systems described in this invention includes:

[0006] (1) Demand-side calculation: Calculate the regulating load based on the load, non-adjustable power output and power receiving data, and calculate the demand-side indicators based on the regulating load. The demand-side indicators include the upward regulating demand, the downward regulating demand, the rate of change and the actual regulating amount.

[0007] (2) Supply-side calculations calculate the regulation capacity of each type of power source according to the power source type, which includes thermal power, hydropower, pumped storage, energy storage and nuclear power;

[0008] (3) The calculated power type regulation capability is aggregated in layers and accumulated step by step to form the inherent, automatic and manual regulation capability of the entire network;

[0009] (4) The calculated regulating load, upward regulating demand and downward regulating demand are used as inputs for the demand item in the regulating margin calculation. The inherent, automatic and manual regulating capacity of the whole network are used as inputs for the capacity item in the regulating margin calculation. The regulating margin is calculated and the regulating margin evaluation result is obtained.

[0010] (5) The mapping and calculation order of the indicator calculation functions required for the initialization steps of demand-side calculation, supply-side calculation, hierarchical aggregation and adjustment margin calculation are initialized through the configuration file, and each step is executed in sequence.

[0011] Furthermore, the regulating load is composed of a weighted combination of the load, non-adjustable power output, and power receiving through configuration coefficients, wherein each configuration coefficient is adapted to a region through a configuration file.

[0012] Furthermore, the upward and downward adjustment demands are extracted from the rising and falling edges of the adjustment load, respectively, and calculated using positive and negative differential methods.

[0013] Furthermore, the rate of change includes the future rate of change and the actual rate of change; the future rate of change is obtained by the difference between the current planned value and the future planned value; the actual rate of change is obtained by the difference between the earliest value and the latest value in the historical data queue.

[0014] Furthermore, the actual adjustment amount is calculated based on historical queue data of the adjustment load, specifically the actual increase adjustment amount and the actual decrease adjustment amount.

[0015] Furthermore, the regulation capacity of thermal power plants is differentiated into automatic frequency regulation mode and peak shaving mode. In automatic frequency regulation mode, the regulation capacity of units with automatic generation control is calculated in segments according to the frequency switching point for future periods. In peak shaving mode, the upward regulation capacity of thermal power units is the positive difference between the planned value and the current active power output, and the downward regulation capacity is the non-positive difference.

[0016] Hydropower regulation capacity is differentiated between the already started and the ready-to-start state. The already started units are calculated based on the current operating conditions, while the ready-to-start units are calculated based on the rated capacity after startup.

[0017] The pumped storage regulation capacity is calculated separately for three operating conditions: power generation, pumping operation, and standby, taking into account water level constraints and planning curve constraints.

[0018] The energy storage regulation capacity is calculated based on the power generation / charging operation status, combined with state of charge constraints and power constraints.

[0019] Nuclear power regulation capacity is calculated based on inherent patterns and does not participate in automatic frequency regulation.

[0020] Further, step (3) includes: first, defining equipment-level and plant-level component tables, binding each indicator to its calculation function, input parameters, plant identifier, and region identifier; second, defining region-level and network-wide total sum tables, establishing the dependency relationship between upper and lower level indicators through the associated indicator identifier; third, constructing a directed dependency graph with each indicator as a node and the indicator pointed to by the associated indicator identifier as an edge, using depth-first search to detect circular dependencies, disconnecting the association between each node on the ring and its parent node after discovering the loop, and performing topological sorting based on in-degree statistics after eliminating the loop; and finally, performing component calculation and total sum aggregation from bottom to top in topological order.

[0021] Furthermore, the calculation of model adjustment margin is divided into automatic upward adjustment margin, automatic downward adjustment margin, manual upward adjustment margin, and manual downward adjustment margin calculation.

[0022] Furthermore, the adjustment margin evaluation results are divided into: normal, early warning, alarm, and emergency; the adjustment margin evaluation results are calculated jointly by automatic upward adjustment margin, automatic downward adjustment margin, manual upward adjustment margin, manual downward adjustment margin, and standby threshold.

[0023] A system for calculating the regulation capacity and fusion statistics of a multi-level power dispatching system includes:

[0024] The multi-source data access layer is used to aggregate four types of data sources: real-time measurement system telemetry data, future planned values ​​from the plan definition table, generator set status information from the generator set table, and automatic generation control mode status.

[0025] The data refresh unit is used to retrieve the latest data from various data sources periodically and maintain a historical data queue and a planned value cache.

[0026] The hierarchical computing engine includes three core sub-modules: loop detection and topology sorting module, component calculation module, and total aggregation module. It is used to construct index dependency graphs and detect cyclic dependencies, perform component calculation and total aggregation, and realize multi-level regulation capability statistics and fusion calculation from equipment to plant, region and the whole network.

[0027] The component calculation module is used to calculate the regulating load based on the load, output of non-adjustable power sources and power receiving data, and to calculate the upward and downward regulation demands on the demand side based on the regulating load; it calculates the regulation capacity of each type of power source, including thermal power, hydropower, pumped storage, energy storage and nuclear power; it uses the inherent, automatic and manual regulation capacity of the entire network as input to calculate the model regulation margin and obtain the regulation margin evaluation results.

[0028] The aggregation module is used to aggregate the power type regulation capabilities obtained by the supply-side calculation module into the inherent, automatic, and manual regulation capabilities of the entire network through hierarchical aggregation.

[0029] The results storage and publishing layer is used to write the calculation results into the component results table and the total sum results table, and generate the upload file.

[0030] The application presentation layer is used to provide a visual monitoring interface for dispatching and operation personnel.

[0031] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: 1. Through hierarchical aggregation, it achieves automatic calculation and statistics at each level based on the hierarchical relationship of the power grid model, reducing the workload of manual maintenance; 2. Integrated monitoring of supply and demand: It incorporates regulation demand (demand side) and the regulation capacity of multiple types of power sources (supply side) into a unified calculation framework, realizing a comprehensive perception of the daily regulation balance of the power grid; 3. Multi-timescale fusion: It simultaneously processes real-time values, historical extreme values, and future planned values, supporting real-time monitoring and ultra-short-term early warning, covering 16 time segments; 4. Flexible and configurable: It can be extended to new areas through multiple algorithm numbering and code; 5. Self-recovery of circular dependencies: Through graph construction, depth-first loop detection and breaking, and topology sorting, it automatically handles complex dependencies between indicators, improving system robustness; 6. Based on the hierarchical evaluation of automatic and manual regulation margins, it provides intuitive risk warning information for dispatching operators, assisting in intraday frequency regulation decisions. Attached Figure Description

[0032] Figure 1 This is a flowchart of the method of the present invention;

[0033] Figure 2 This is a diagram of the collaborative computing framework for the supply and demand sides of this invention;

[0034] Figure 3 This is a diagram of the hierarchical aggregation calculation framework of the present invention;

[0035] Figure 4 This is a diagram of the computational scheduling and multi-algorithm adaptation framework of the present invention;

[0036] Figure 5 This is the overall system architecture diagram of the present invention. Detailed Implementation

[0037] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0038] The method for fusion calculation of regulation capacity for multi-level power dispatching systems described in this invention includes:

[0039] (1) Demand-side calculation: Calculate the regulating load based on the load, non-adjustable power output and power receiving data, and calculate the demand-side indicators based on the regulating load. The demand-side indicators include the upward regulating demand, the downward regulating demand, the rate of change and the actual regulating amount.

[0040] The regulating load is a weighted combination of the load, the output of the non-adjustable power source, and the power receiving source, calculated using a configuration factor. The formula for calculating the regulating load is:

[0041]

[0042] Each coefficient is adapted to a specific region through a configuration file;

[0043] The upward and downward adjustment demands are extracted from the rising and falling edges of the adjustment load, respectively, and calculated using positive and negative differential methods, as follows:

[0044]

[0045]

[0046] In the formula, This represents the adjustment load over 15 minutes. This represents the current adjustment load.

[0047] The future rate of change is obtained by the difference between the current planned value and the future planned value; the actual rate of change is calculated by the difference between the earliest and latest values ​​in the historical data queue (T-15 minutes). The actual adjustment amount is calculated based on the historical queue data of the regulated load, specifically the actual upward adjustment amount and the actual downward adjustment amount.

[0048] (2) Supply-side calculations calculate the regulation capacity of each type of power source according to the power source type, which includes thermal power, hydropower, pumped storage, energy storage and nuclear power;

[0049] See Figure 2 Supply-side regulation capacity is categorized into five types based on power source: thermal power, hydropower, pumped storage, energy storage, and nuclear power. It is also categorized by regulation method: inherent regulation capacity (peak-shaving power sources contribute naturally according to plan), automatic regulation capacity (units that participate in frequency regulation through automatic generation control can automatically adjust their regulation capacity), and manual regulation capacity (dispatch and adjustment capacity can be intervened by dispatching personnel). Thermal power, hydropower, pumped storage, and energy storage all possess inherent, automatic, and manual regulation capabilities; nuclear power only provides inherent regulation capacity and does not participate in automatic frequency regulation.

[0050] Thermal power regulation capacity is differentiated between automatic frequency regulation (AFM) and peak shaving modes, each employing different calculation strategies. In AFM mode, units with automatic generation control are calculated for future time periods segmented by frequency switching points. In peak shaving mode, the upward regulation capacity of thermal power units is the positive difference between the planned value and the current active power output, while the downward regulation capacity is a non-positive difference. Hydropower regulation capacity is differentiated between operational and ready-to-operate states. Operationally operational units are calculated based on their current operating conditions, while ready-to-operate units are calculated based on their rated capacity after startup. Pumped storage regulation capacity is calculated separately for three operating conditions: generation operation, pumping operation, and standby, taking into account water level constraints and planned curve constraints. Energy storage regulation capacity is calculated based on the generation / charging operation status, combined with state-of-charge constraints and power constraints. Nuclear power regulation capacity is calculated using its inherent mode and does not participate in AFM.

[0051] After the regulation capacity of various power sources on the supply side is calculated, component calculations are performed at the equipment level, and then aggregated level by level at the plant, regional, and whole network levels to form the inherent, automatic, and manual regulation capacity of thermal power, hydropower, pumped storage, and energy storage, as well as the inherent regulation capacity of nuclear power.

[0052] (3) The calculated power type regulation capability is aggregated in layers and accumulated step by step to form the inherent, automatic and manual regulation capability of the entire network;

[0053] See Figure 3 Supply-side computation can be represented as a hierarchical aggregation process: First, define equipment-level and plant-level component tables, binding each indicator to its calculation function, input parameters, plant identifier, and region identifier; second, define region-level and network-wide aggregate tables, establishing dependencies between upper and lower level indicators through associated indicator identifiers; third, construct a directed dependency graph with each indicator as a node and the indicator pointed to by the associated indicator identifier as an edge, using depth-first search to detect circular dependencies, disconnecting the association between each node on the loop and its parent node after a loop is found, and performing topological sorting based on in-degree statistics after eliminating the loop; finally, perform component calculation and aggregate aggregation from bottom to top in topological order; the component tables correspond to equipment-level / plant-level component calculations, with the data source type being "calculation," and binding the calculation functions and input parameters of each indicator through configuration files; the aggregate tables correspond to region-level / network-wide automatic aggregation, with the data source type being "statistics," and achieving flexible multi-level aggregation based on the collaboration of associated indicator identifiers and region identifiers.

[0054] During aggregation, the system works in conjunction with association indicator identifiers and region identifiers: association indicator identifiers are used to establish dependencies between indicators, construct a directed graph, and determine the calculation order; during aggregation, the results of lower-level indicators are accumulated to the results of higher-level indicators by matching association indicator identifiers. Region identifiers are used to define the regional scope to which an indicator belongs; during aggregation, the results of all lower-level indicators within the same region are accumulated by matching region identifiers. The coordinated configuration of these two identifiers enables the system to support complex hierarchical aggregation scenarios.

[0055] Building upon this foundation, the system supports two extended aggregation modes. The first is a hybrid aggregation mode: a region can contain both the calculation results of its directly subordinate plants and stations, as well as the aggregation results of its lower-level sub-regions. For example, region A is aggregated from plants A1, plants A2, and sub-region A1. The second is a cross-region aggregation mode: a region can be aggregated from the results of multiple other regions at the same level. For example, region D is aggregated from the results of regions B and C. Both modes are implemented through the joint configuration of associated indicator identifiers and region identifiers, requiring no modification to the program code.

[0056] The core execution flow of hierarchical aggregation is as follows: Equipment-level data (such as real-time measurement and planned data of thermal power units, hydropower units, wind power units, energy storage units, and pumped-storage units) are first calculated in the component table, and the calculation results are cached in the component calculation result cache area; at the plant level, the components of each equipment within the same plant are accumulated; at the sub-region level, after completing internal aggregation, the results are output to the higher-level region; at the region level, the component results of directly subordinate plants and lower-level sub-regions are accumulated through joint matching of associated indicator identifiers and region identifiers; at the network-wide level, the components of each region are accumulated. The total aggregation result is cached in the total aggregation result cache area. For total aggregation indicators with a data source type of "calculation", the system automatically triggers the total aggregation of the statistical type of that indicator after the component calculation is completed, without manual intervention.

[0057] The system supports the calculation and aggregation of 65 types of indicators, covering demand-side adjustment demand indicators (adjustment load, upward and downward adjustment demand, rate of change, actual adjustment amount, peak-valley difference, etc.), supply-side power source adjustment capacity indicators (inherent and automatic and manual adjustment capacity of thermal power, hydropower, pumped storage, energy storage, nuclear power), network-wide summary indicators (inherent, automatic, and manual upward and downward adjustment capacity of the entire network and extreme output values ​​of various power sources, etc.), and margin evaluation indicators (margin, threshold, four-level evaluation results, etc.), comprehensively covering all kinds of statistical quantities required for intraday adjustment monitoring.

[0058] (4) The calculated regulating load, upward regulating demand and downward regulating demand are used as inputs for the demand item in the regulating margin calculation. The inherent, automatic and manual regulating capacity of the whole network are used as inputs for the capacity item in the regulating margin calculation. The regulating margin is calculated and the regulating margin evaluation result is obtained.

[0059] After calculating the overall network regulation capacity and regulation demand, the regulation margin is calculated, which is equal to the regulation capacity minus the regulation demand.

[0060] Model adjustment margin calculation is divided into automatic upward adjustment margin, automatic downward adjustment margin, manual upward adjustment margin, and manual downward adjustment margin calculation.

[0061] The formula for calculating the automatic adjustment margin is:

[0062]

[0063] The formula for calculating the automatic adjustment margin is:

[0064]

[0065] The formula for calculating the manual adjustment margin is:

[0066]

[0067] The formula for calculating the manual adjustment margin is:

[0068]

[0069] The subscript T in the formula represents the current time.

[0070] The adjustment margin evaluation results are categorized as: Normal, Warning, Alarm, and Emergency. The adjustment margin evaluation result is calculated jointly by the automatic upward adjustment margin, automatic downward adjustment margin, manual upward adjustment margin, manual downward adjustment margin, and the reserve threshold. Specifically, a normal condition is defined as both the automatic upward and downward adjustment margins not less than the reserve threshold; a warning condition is defined as both the automatic upward and downward adjustment margins not less than zero and at least one direction being less than the reserve threshold; an alarm condition is defined as either the automatic upward or downward adjustment margin being less than zero and the corresponding manual margin not less than zero; and an emergency condition is defined as either the manual upward or downward adjustment margin being less than zero. In other words:

[0071] Normal: Automatic upward adjustment margin ≥ standby threshold and automatic downward adjustment margin ≥ standby threshold;

[0072] Warning: Automatic upward adjustment margin ≥ 0 and automatic downward adjustment margin ≥ 0 and (automatic upward adjustment margin < reserve threshold or automatic downward adjustment margin < reserve threshold);

[0073] Alarm: (Automatic upward adjustment margin < 0 or automatic downward adjustment margin < 0) and manual upward adjustment margin ≥ 0 and manual downward adjustment margin ≥ 0;

[0074] Emergency: Manually increase adjustment margin < 0 or manually decrease adjustment margin < 0;

[0075] in, .

[0076] The evaluation results are stored in the form of integer status codes, supporting simultaneous evaluation of two time scales: real-time values ​​and future planned values ​​(T+15 to T+120), providing dispatching and operation personnel with risk warning information for the entire process within the day.

[0077] (5) Computation scheduling and multi-algorithm adaptation: The index calculation function mapping and calculation order required for demand-side calculation, supply-side calculation, hierarchical aggregation and adjustment margin calculation are initialized through configuration file, and each step is driven to be executed in sequence.

[0078] See Figure 4 The system uses a configuration file-driven approach to initialize the calculation parameters for each indicator. The parameter manager is responsible for establishing the calculation function mapping and calculation order for 65 types of indicators, and specifies the calculation rules used in the current region through the algorithm number parameter (value range 1 to 7) in the configuration file.

[0079] To address the complex dependencies between indicators in the summary table, the system employs directed graph modeling: each indicator is treated as a node, and the associated indicators pointed to by `result_id` are used as edges to construct an indicator dependency graph. A depth-first search is used to detect circular dependencies; if a cycle is found, the association between each node in the cycle and its parent node is broken, eliminating the cycle. Subsequently, a topological sort is performed based on in-degree statistics to determine the bottom-up calculation order. This mechanism automates the handling of complex dependencies between indicators, improving the system's robustness.

[0080] The computation scheduling adopts a phased execution strategy: under the default algorithm, the component calculation and total aggregation of all indicators are executed sequentially in the first phase; for specific algorithms (such as Algorithm 7), the system automatically divides the phase at the related indicators. In each phase, the component calculation of all statistical types within the phase is executed first, and then the corresponding total aggregation is executed, ensuring that downstream indicators are calculated in the downstream phase and upstream indicators are completed in the upstream phase first, avoiding cross-phase dependency conflicts.

[0081] The system supports dynamic adaptation of multiple algorithms. By selecting different regional rules through the algorithm number parameter (value range 1 to 7) in the configuration file, the same program can be adapted to the differentiated calculation rules of different provincial or regional power grids, and can be extended to new regions without modifying the program code. Typical differentiated rules include: whether thermal power units filter automatic frequency regulation mode, whether the output of self-owned power plants and local power plants is deducted when regulating load, whether hydropower is distinguished between already started and ready to start, and whether the thermal power capacity in future time periods is segmented according to the frequency regulation switching point, etc.

[0082] The overall architecture of the regulation capability and fusion statistical calculation system for multi-level power dispatching systems described in this invention is divided into five layers: including:

[0083] The multi-source data access layer is used to aggregate four types of data sources: telemetry data from the real-time measurement system, future planned values ​​from the plan definition table (including 300-second acquisition boundaries), generator set status information from the generator set table, and automatic generation control mode status.

[0084] The data refresh unit is used to retrieve the latest data from various data sources periodically and maintain a historical data queue (T-15 minutes) and a planned value cache.

[0085] The hierarchical computing engine includes three core sub-modules: loop detection and topology sorting module, component calculation module, and total aggregation module. It is used to construct index dependency graphs and detect cyclic dependencies, perform component calculation and total aggregation, and realize multi-level regulation capability statistics and fusion calculation from equipment to plant, region and the whole network.

[0086] The component calculation module is used to calculate the regulating load based on the load, output of non-adjustable power sources and power receiving data, and to calculate the upward and downward regulation demands on the demand side based on the regulating load; it calculates the regulation capacity of each type of power source, including thermal power, hydropower, pumped storage, energy storage and nuclear power; it uses the inherent, automatic and manual regulation capacity of the entire network as input to calculate the model regulation margin and obtain the regulation margin evaluation results.

[0087] The aggregation module is used to aggregate the power type regulation capabilities obtained by the supply-side calculation module into the inherent, automatic, and manual regulation capabilities of the entire network through hierarchical aggregation.

[0088] The results storage and publishing layer is used to write the calculation results into the component results table and the total sum results table, and generate the upload file.

[0089] The application presentation layer provides a visual monitoring interface for dispatching and operation personnel.

[0090] The system starts calculation tasks at fixed intervals, with a processing cycle of 10 seconds and a result file generation cycle of 15 minutes. It simultaneously processes real-time values, historical extreme values, and future planned values ​​(from T+15 minutes to T+120 minutes), covering the calculation results of 16 time segments, of which 17 are planned value segments (including the current time).

Claims

1. A method for fusion calculation of regulation capacity for multi-level power dispatching systems, characterized in that, include: (1) Demand-side calculation: Calculate the regulating load based on the load, non-adjustable power output and power receiving data, and calculate the demand-side indicators based on the regulating load. The demand-side indicators include the upward regulating demand, the downward regulating demand, the rate of change and the actual regulating amount. (2) Supply-side calculation: The regulation capacity of each type of power source is calculated according to the power source type, which includes thermal power, hydropower, pumped storage, energy storage and nuclear power. (3) The calculated power type regulation capabilities are aggregated in layers and accumulated step by step to form the inherent, automatic, and manual regulation capabilities of the entire network; (4) The calculated regulating load, upward regulating demand and downward regulating demand are used as inputs for the demand item in the regulating margin calculation. The inherent, automatic and manual regulating capacity of the whole network are used as inputs for the capacity item in the regulating margin calculation. The regulating margin is calculated and the regulating margin evaluation result is obtained. (5) The mapping and calculation order of the indicator calculation functions required for the initialization steps of demand-side calculation, supply-side calculation, hierarchical aggregation and adjustment margin calculation are initialized through the configuration file, and each step is executed in sequence.

2. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, The regulating load is composed of the load, non-adjustable power output and power receiving through a weighted combination of configuration coefficients, where each configuration coefficient is adapted to a region through a configuration file.

3. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, The upward and downward adjustment demands are extracted from the rising and falling edges of the adjustment load, respectively, and calculated using positive and negative differential methods.

4. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, The rate of change includes the future rate of change and the actual rate of change; the future rate of change is obtained by the difference between the current planned value and the future planned value; the actual rate of change is obtained by the difference between the earliest value and the latest value in the historical data queue.

5. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, The actual adjustment amount is calculated based on historical queue data of the adjustment load, specifically the actual increase adjustment amount and the actual decrease adjustment amount.

6. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, Thermal power regulation capacity is differentiated into automatic frequency regulation mode and peak shaving mode. In automatic frequency regulation mode, the regulation capacity of units with automatic generation control is calculated in segments according to the frequency switching point for future time periods. In peak shaving mode, the upward regulation capacity of thermal power units is the positive difference between the planned value and the current active power output, and the downward regulation capacity is the non-positive difference. Hydropower regulation capacity is differentiated between the already started and the ready-to-start state. The calculated capacity of the already started units is based on the current operating conditions, while the calculated capacity of the ready-to-start units is based on the rated capacity after startup. The pumped storage regulation capacity is calculated separately for three operating conditions: power generation, pumping operation, and standby, taking into account water level constraints and planning curve constraints. The energy storage regulation capacity is calculated based on the power generation / charging operation status, combined with state of charge constraints and power constraints. Nuclear power regulation capacity is calculated based on inherent patterns and does not participate in automatic frequency regulation.

7. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, Step (3) includes: First, defining equipment-level and plant-level component tables, binding each indicator to its calculation function, input parameters, plant identifier, and region identifier; second, defining region-level and network-wide total sum tables, establishing the dependency relationship between upper and lower level indicators through associated indicator identifiers; third, constructing a directed dependency graph with each indicator as a node and the indicator pointed to by the associated indicator identifier as an edge, using depth-first search to detect circular dependencies, disconnecting the association between each node on the ring and its parent node after discovering the loop, and performing topological sorting based on in-degree statistics after eliminating the loop; finally, performing component calculation and total sum aggregation from bottom to top in topological order.

8. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, Model adjustment margin calculation is divided into automatic upward adjustment margin, automatic downward adjustment margin, manual upward adjustment margin, and manual downward adjustment margin calculation.

9. The method for calculating the fusion of adjustment capabilities according to claim 1, characterized in that, The adjustment margin evaluation results are divided into: normal, early warning, alarm, and emergency. The adjustment margin evaluation results are calculated by automatically increasing the adjustment margin, automatically decreasing the adjustment margin, manually increasing the adjustment margin, manually decreasing the adjustment margin, and the standby threshold.

10. A system for calculating the regulation capacity and fusion statistics of a multi-level power dispatching system, characterized in that, include: The multi-source data access layer is used to aggregate four types of data sources: real-time measurement system telemetry data, future planned values ​​from the plan definition table, generator set status information from the generator set table, and automatic generation control mode status. The data refresh unit is used to retrieve the latest data from various data sources periodically and maintain a historical data queue and a planned value cache. The hierarchical computing engine includes three core sub-modules: loop detection and topology sorting module, component calculation module, and total aggregation module. It is used to construct index dependency graphs and detect cyclic dependencies, perform component calculation and total aggregation, and realize multi-level regulation capability statistics and fusion calculation from equipment to plant, region and the whole network. The component calculation module is used to calculate the regulating load based on the load, output of non-adjustable power sources and power receiving data, and to calculate the upward and downward regulation demands on the demand side based on the regulating load; it calculates the regulation capacity of each type of power source, including thermal power, hydropower, pumped storage, energy storage and nuclear power; it uses the inherent, automatic and manual regulation capacity of the entire network as input to calculate the model regulation margin and obtain the regulation margin evaluation results. The aggregation module is used to aggregate the power type regulation capabilities obtained by the supply-side calculation module into the inherent, automatic, and manual regulation capabilities of the entire network through hierarchical aggregation. The results storage and publishing layer is used to write the calculation results into the component results table and the total sum results table, and generate the upload file. The application presentation layer provides a visual monitoring interface for dispatching and operation personnel.

Images