AGC coordinated optimization control system based on multi-source data fusion

Abstract

The invention relates to the technical field of AGC coordinated control, and discloses an AGC coordinated optimization control system based on multi-source data fusion, and the system carries out the collection and normalization preprocessing of multi-source data through a multi-source data collection module, and forms a global multi-dimensional data set; constructing a risk model through a data driving module, outputting a minimum reserve capacity demand adaptive to a current system risk level, and determining safety demands and economic boundaries under different working conditions through working condition discrimination and economic cost benchmark calculation; a hierarchical multi-objective optimization model is constructed through an intelligent optimization module, and a resource allocation scheme considering a safety base line and economic optimization is solved; and disassembling the optimal decision scheme into personalized instructions adaptive to various resource physical characteristics through a personalized adjustment module. And finally, an AGC coordination optimization mechanism with dynamic balance of safety and economy is constructed, and the problem of contradiction between standby capacity guarantee safety and control cost is solved accurately.

Description

Technical Field

[0001] This invention relates to the field of AGC coordination control technology, specifically to an AGC coordination optimization control system based on multi-source data fusion. Background Technology

[0002] AGC (Automatic Generation Control) is a key technology in power systems used to automatically adjust generator output to maintain grid frequency stability and ensure that tie-line power operates according to planned values. It monitors parameters such as grid frequency and tie-line power in real time, compares these with setpoints, calculates the required adjustment in generator output, and issues commands to each generator unit.14 In actual operation, grid load changes constantly, such as during peak daytime electricity consumption and during off-peak nighttime consumption. AGC ensures that generator units respond promptly to these changes, avoiding grid frequency fluctuations and guaranteeing the safe and stable operation of the power system.1 Simultaneously, this control technology can dynamically optimize the combination and operation mode of generator units based on grid load conditions, enabling units to operate under high-efficiency conditions, reducing generation costs, improving energy utilization efficiency, and achieving economical operation of the power system.

[0003] To ensure frequency stability, sufficient reserve capacity needs to be reserved, but reserve capacity will increase power generation costs. If economic optimization is pursued, the reserve margin may be reduced, resulting in a decrease in the system's ability to withstand disturbances. The current AGC coordinated optimization control system is unable to find a suitable balance between safety priority and economic optimization. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides an AGC coordinated optimization control system based on multi-source data fusion. It features data-driven quantitative decision-making, hierarchical optimization, and personalized execution, constructing an AGC coordinated optimization mechanism that dynamically balances safety and economy. This precisely solves the problem of the contradiction between backup capacity guaranteeing safety and control costs. It ensures the backup margin required for system frequency stability while minimizing the power generation costs caused by redundant backup, ultimately finding the optimal solution for dynamic balance between safety priority and economic optimization.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution: an AGC coordination and optimization control system based on multi-source data fusion, comprising a multi-source data acquisition module, a data driving module, an intelligent optimization module, and a personalized adjustment module;

[0008] The multi-source data acquisition module is used to collect power grid operation data, regulation resource data, market transaction data, and risk scenario data, and performs normalization preprocessing on the collected data to form a full-domain multi-dimensional data set.

[0009] The data-driven module introduces conditional value of risk indicators based on a full-domain multi-dimensional data set to construct a risk model, outputs minimum standby capacity requirements, simultaneously identifies operating condition types, and calculates the economic cost benchmark under different operating conditions.

[0010] The intelligent optimization module uses minimum reserve capacity as a hard constraint and economic cost benchmark as the objective, integrates a multi-dimensional dataset across the entire domain to construct a hierarchical multi-objective optimization model, and solves for the optimal decision scheme.

[0011] The personalized adjustment module, based on resource characteristic data in the multi-dimensional data of the whole domain, decomposes the optimal decision scheme into personalized instruction outputs that adapt to each resource. Each regional scheduling node receives the personalized instructions and executes the processing adjustment instructions.

[0012] Preferably, the data-driven module selects core data that are strongly related to risk assessment, working condition identification, and cost calculation from the multi-dimensional dataset, specifically including risk association dataset, working condition identification dataset, and cost calculation dataset;

[0013] The risk-related dataset includes unit tripping power, sudden drop in renewable energy, load mutation, renewable energy penetration rate, unit health, and tie line transmission margin.

[0014] The operating condition discrimination dataset includes real-time frequency, rated frequency, real-time load, and maximum load;

[0015] The cost calculation dataset includes conventional unit resource types, energy storage resource types, market resource types, minimum output limits for each type of resource, maximum output limits for each type of resource, and unit regulation cost parameters.

[0016] Preferably, in the data-driven module, the risk model expression is: ; in, This represents the output of the risk model, namely the minimum reserve capacity requirement; Represents the core items of risk quantification. The power deficit random variables represent the power of unit tripping, the magnitude of the sudden drop in renewable energy, and the load change, corresponding to each single disturbance scenario. Represents a risk quantification coefficient; Represents the risk deviation correction term; The weighting coefficient represents the penetration rate of new energy sources; Represents the penetration rate of new energy sources; Represents the weighting coefficient of unit health; Represents the health status of the generator set; Represents the transmission margin weighting coefficient of the tie line; This represents the transmission margin of the tie line.

[0017] Preferably, the data-driven module calculates the frequency deviation and load rate based on the operating condition discrimination dataset to determine the operating condition type. The calculation formulas for the frequency deviation and load rate are as follows: ; ; in, Represents frequency deviation; Represents real-time frequency; Represents the rated frequency; Represents load factor; Represents real-time load; This represents the maximum load.

[0018] Preferably, the method for determining the type of working condition is as follows:

[0019] when , representing the disturbance condition; when and This represents peak operating conditions; when and This represents a low-end operating condition; when and , representing the flat state operating condition.

[0020] Preferably, in the data-driven module, the economic cost benchmark formula expression under different operating conditions is as follows: ; in, Representing the Economic cost benchmark for similar operating conditions; Represents the type of resource to be regulated. Represents conventional generating units. Represents energy storage. Represents market resources; Representing the Class resources in the Unit adjustment cost under similar operating conditions; This represents the maximum acceptable unit cost of backup. Represents an indicator function; This means that only under disturbed or peak operating conditions, the cost of additional backup redundancy is added. This represents the cost of backup redundancy.

[0021] Preferably, in the intelligent optimization module, the hierarchical multi-objective optimization model is expressed by the following formula: ; ; in, This represents the objective function to be optimized; Representing the Within the scheduling period, the first Class resources in the The optimal decision-making scheme under similar working conditions, i.e., the output result of the hierarchical multi-objective optimization model; Representing the Total system reserve capacity during each scheduling period; , Represents weight;

[0022] Represents constraints; This represents a core security constraint, mandating that the total system backup capacity must not be less than the minimum backup capacity; This means that the optimal capacity allocation is limited between the minimum and maximum output limits. Representing the Class resources in the Minimum output limit under similar working conditions Representing the Class resources in the Maximum output limit under similar working conditions.

[0023] Preferably, the The calculation formula is: .

[0024] Preferably, the personalized adjustment module extracts all resources from the full-domain multi-dimensional data set. The characteristic parameters are categorized by resource type:

[0025] conventional units Minimum start-up and shutdown time, ramp rate, coal consumption curve, and peak shaving range;

[0026] Energy storage Charge / discharge efficiency, capacity limit, response time delay, and cycle life constraints;

[0027] Market Resources Price quote curve, call response time, and adjustable capacity range;

[0028] Match the working condition type of the current scheduling period Retrieve the threshold values ​​of resource characteristic parameters under the corresponding operating conditions.

[0029] Preferably, the personalized adjustment module outputs the hierarchical multi-objective optimization model. By splitting by resource type, the target allocated capacity of a single resource under the current operating conditions is obtained and marked as follows. Then to Perform hard constraint verification; if the physical limits are exceeded, correct according to the rules:

[0030] like ,but The difference is allocated to other resources according to the principle of optimal cost.

[0031] like ,but Any excess will be included in the system's backup redundancy.

[0032] like ,but ;

[0033] Representing the Within the scheduling period, the first Class resources in the Final executable allocation capacity under similar operating conditions; the corrected Convert the command parameters into executable parameters for each resource:

[0034] Conventional generating units: converted to output adjustment commands; Energy storage: converted to charge / discharge mode commands; Market resources: converted to capacity call commands.

[0035] Compared with existing technologies, this invention provides an AGC coordinated optimization control system based on multi-source data fusion, which has the following beneficial effects:

[0036] This invention constructs an AGC (Automatic Gain Control) coordination and optimization mechanism that dynamically balances safety and economy through data-driven quantitative decision-making, hierarchical optimization, and personalized execution. It precisely addresses the challenge of balancing reserve capacity security with cost control. A multi-source data acquisition module collects and normalizes multi-source data, forming a comprehensive multi-dimensional data set covering power grid operation data, regulation resource data, market transaction data, and risk scenario data. This provides high-quality, consistent data support for all subsequent stages. A data-driven module builds a risk model based on this comprehensive multi-dimensional data set, outputting the minimum reserve capacity requirement adapted to the current system risk level. Simultaneously, it uses operating condition judgment and economic... Cost benchmark calculation clarifies the safety requirements and economic boundaries under different operating conditions, defining necessary reserves rather than excessive reserves from the source. Through the intelligent optimization module, a hierarchical multi-objective optimization model is constructed with minimum reserve capacity as a hard constraint and economic cost benchmark as the optimization objective. By flexibly adjusting the priority of safety and economy through weight coefficients, a resource allocation scheme that balances safety bottom line and economic optimization is obtained. Through the personalized adjustment module, the optimal decision scheme is decomposed into personalized instructions adapted to the physical characteristics of various resources. After hard constraint verification and differential redistribution, it is ensured that the instructions are executable and risk-free. Finally, through parallel execution by scheduling nodes in various regions, the precise implementation from model decision to resource adjustment is achieved.

[0037] This approach ensures the reserve margin required for system frequency stability while minimizing the power generation costs associated with redundant reserves, ultimately finding an optimal solution that achieves a dynamic balance between prioritizing safety and optimizing economics. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It is worth noting that this application also relates to prior art. Since prior art is well known to those skilled in the art, it will not be described in detail in this application.

[0040] Please see Figure 1 The AGC coordination and optimization control system based on multi-source data fusion includes a multi-source data acquisition module, a data-driven module, an intelligent optimization module, and a personalized adjustment module.

[0041] The multi-source data acquisition module is used to collect power grid operation data, regulation resource data, market transaction data, and risk scenario data. It performs normalization preprocessing on the collected data using linear normalization to form a multi-dimensional dataset covering the entire domain. This breaks down data silos, unifies data dimensions and value ranges, and provides high-quality, highly consistent data support for subsequent risk model construction, operating condition identification, cost benchmark calculation, and hierarchical multi-objective optimization. At the same time, the integration of multi-source data can comprehensively cover the entire link requirements of AGC coordinated optimization control. Normalization preprocessing eliminates the differences in dimensions between different data, avoids unreasonable weight bias of parameters with large differences in numerical range on the model calculation results, and improves the convergence speed and calculation accuracy of the model. Ultimately, it ensures the scientific nature and feasibility of the optimal decision-making scheme.

[0042] The data-driven module selects core data that are strongly related to risk assessment, operating condition identification, and cost calculation from the multi-dimensional dataset across the entire domain to build a risk model and output the minimum standby capacity requirement. The core data includes risk correlation dataset, operating condition identification dataset, and cost calculation dataset.

[0043] The risk-related dataset includes unit tripping power, sudden drop in renewable energy, load mutation, renewable energy penetration rate, unit health, and tie line transmission margin.

[0044] The operating condition discrimination dataset includes real-time frequency, rated frequency, real-time load, and maximum load;

[0045] The cost calculation dataset includes conventional unit resource types, energy storage resource types, market resource types, minimum output limits for each type of resource, maximum output limits for each type of resource, and unit regulation cost parameters;

[0046] The risk model expression is: ; in, This represents the output of the risk model, namely the minimum reserve capacity requirement; Represents the core items of risk quantification. The power deficit random variables represent the power of unit tripping, the magnitude of the sudden drop in renewable energy, and the load change, corresponding to each single disturbance scenario. Represents a risk quantification coefficient; Represents the risk deviation correction term; The weighting coefficient represents the penetration rate of new energy sources; Represents the penetration rate of new energy sources; Represents the weighting coefficient of unit health; Represents the health status of the generator set; Represents the transmission margin weighting coefficient of the tie line; Represents the transmission margin of the tie line;

[0047] The risk model expression quantifies multi-dimensional risk factors and outputs a precise minimum reserve capacity requirement. This provides a core quantitative basis for balancing the contradiction between safety and economy in the AGC coordinated optimization control system. It abandons the extensive mode of empirical fixed reserve capacity in traditional AGC control, and couples and quantifies the power deficit in random disturbance scenarios such as unit tripping, sudden drop in renewable energy, and load change with system operating status parameters such as renewable energy penetration rate, unit health, and tie line transmission margin. At the same time, it introduces a deviation correction term and finally outputs the minimum necessary reserve capacity that adapts to the current system risk level. This minimum necessary reserve capacity is embedded as a hard constraint in the subsequent hierarchical multi-objective optimization model to ensure that the system reserve capacity is sufficient to resist the current risk, ensure frequency stability, and avoid the decline in anti-disturbance capability due to insufficient reserve. It clarifies the minimum necessary reserve boundary and eliminates the waste of redundant reserve from the source. This lays the foundation for subsequent calculation of the economic cost benchmark for different working conditions and optimization of resource allocation schemes.

[0048] The data-driven module calculates frequency deviation and load rate based on the operating condition discrimination dataset to determine the operating condition type. The formulas for calculating frequency deviation and load rate are as follows: ; ; in, Represents frequency deviation; Represents real-time frequency; Represents the rated frequency; Represents load factor; Represents real-time load; Represents the maximum load;

[0049] when , representing the disturbance condition; when and This represents peak operating conditions; when and This represents a low-end operating condition; when and This represents a flat operating condition;

[0050] By using quantitative thresholds for two core indicators, frequency deviation and load factor, the system achieves accurate and rapid classification of power grid operating conditions. This provides a condition-differentiated decision-making basis for the AGC coordinated optimization control system to balance the contradiction between safety and economy, enabling AGC coordinated optimization to shift from static unified control to dynamic condition-specific optimization. This allows for precise control of the balance between safety and economy under different operating conditions, effectively solving the problem of the difficulty in achieving both.

[0051] The data-driven module calculates the economic cost benchmark for different operating conditions based on the operating condition type. The formula expression is as follows: ; in, Representing the Economic cost benchmark for similar operating conditions; Represents the type of resource to be regulated. Represents conventional generating units. Represents energy storage. Represents market resources; Representing the Class resources in the Unit adjustment cost under similar operating conditions; This represents the maximum acceptable unit cost of backup. Represents an indicator function; This means that only under disturbed or peak operating conditions, the cost of additional backup redundancy is added. Represents the cost of backup redundancy;

[0052] The economic cost benchmark calculation formula under different operating conditions outputs an economic cost benchmark that is adapted to different power grid operating conditions through quantitative accounting by operating condition and resource type. It anchors the direction of economically optimal resource allocation from the source. It accurately distinguishes the differences in operating conditions through indicator functions, and only adds the backup redundancy cost under disturbance or peak operating conditions. This avoids the waste of redundant costs under low-temperature and flat operating conditions, and ensures the rationality of safety investment under high-risk operating conditions.

[0053] The intelligent optimization module uses minimum reserve capacity as a hard constraint and economic cost as the objective. It integrates a multi-dimensional dataset across the entire domain to construct a hierarchical multi-objective optimization model and solves for the optimal decision-making scheme. The formula for the hierarchical multi-objective optimization model is as follows: ; ; ; in, This represents the objective function to be optimized; Representing the Within the scheduling period, the first Class resources in the The optimal decision-making scheme under similar working conditions, i.e., the output result of the hierarchical multi-objective optimization model; Representing the Total system reserve capacity during each scheduling period; , Represents weight;

[0054] Represents constraints; This represents a core security constraint, mandating that the total system backup capacity must not be less than the minimum backup capacity; This means that the optimal capacity allocation is limited between the minimum and maximum output limits. Representing the Class resources in the Minimum output limit under similar working conditions Representing the Class resources in the Maximum output limit under similar working conditions;

[0055] The hierarchical multi-objective optimization model takes the lower limit constraint of the total system reserve capacity and the physical output boundary of resources as rigid premises, fundamentally eliminating the risk of sacrificing the system's disturbance resistance in pursuit of economy. The model directly outputs the optimal allocation capacity of each resource under different time periods and operating conditions. This enables precise implementation from balancing macro-level goals to scheduling micro-level resources, allowing AGC coordination optimization to shift from experience-based decision-making to data-driven quantitative decision-making.

[0056] The personalized adjustment module extracts all resources from the multi-dimensional dataset across the entire domain. The characteristic parameters are categorized by resource type:

[0057] conventional units Minimum start-up and shutdown time, ramp rate, coal consumption curve, and peak shaving range;

[0058] Energy storage Charge / discharge efficiency, capacity limit, response time delay, and cycle life constraints;

[0059] Market Resources Price quote curve, call response time, and adjustable capacity range;

[0060] Match the working condition type of the current scheduling period Retrieve the threshold values ​​of resource characteristic parameters under the corresponding working conditions;

[0061] The personalized adjustment module will output the hierarchical multi-objective optimization model. By splitting by resource type, the target allocated capacity of a single resource under the current operating conditions is obtained and marked as follows. Then to Perform hard constraint verification; if the physical limits are exceeded, correct according to the rules:

[0062] like ,but The difference is allocated to other resources according to the principle of optimal cost.

[0063] like ,but Any excess will be included in the system's backup redundancy.

[0064] like ,but ;

[0065] Representing the Within the scheduling period, the first Class resources in the Final executable allocation capacity under similar operating conditions; the corrected Convert the command parameters into executable parameters for each resource:

[0066] Conventional generating units: converted to output adjustment commands; Energy storage: converted to charge / discharge mode commands; Market resources: converted to capacity allocation commands;

[0067] Each regional dispatch node receives personalized instructions and executes processing and adjustment instructions;

[0068] The personalized adjustment module transforms the theoretical decision-making schemes output by the hierarchical multi-objective optimization model into executable instructions adapted to the physical characteristics of various resources. By extracting and matching the characteristic parameters and operating thresholds of different resources, it ensures that the instructions comply with the start-up and shutdown constraints and ramp-up constraints of conventional units, the capacity and lifespan constraints of energy storage, and the adjustable range and response constraints of market resources. This avoids invalid instructions that exceed the operating limits of the equipment. Through hard constraint verification and differential redistribution rules, it maximizes the retention of economic optimization effects while ensuring that the total reserve capacity of the system meets the safety hard constraints. This eliminates the risk of resource overload and reduces the waste of redundant reserves. The module differentiates the instruction form according to resource type, directly mapping the final allocated capacity to executable operations such as unit output adjustment, energy storage charging and discharging mode, and market resource call. This significantly improves the instruction execution efficiency and accuracy of the AGC coordinated optimization control system, ensuring that the balance between safety and economy is truly realized in the operation and control of each type of adjustable resource.

[0069] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An AGC coordinated optimization control system based on multi-source data fusion, characterized in that, It includes a multi-source data acquisition module, a data-driven module, an intelligent optimization module, and a personalized adjustment module; The multi-source data acquisition module is used to collect power grid operation data, regulation resource data, market transaction data, and risk scenario data, and performs normalization preprocessing on the collected data to form a full-domain multi-dimensional data set. The data-driven module introduces conditional value of risk indicators based on a full-domain multi-dimensional data set to construct a risk model, outputs minimum standby capacity requirements, simultaneously identifies operating condition types, and calculates the economic cost benchmark under different operating conditions. The intelligent optimization module uses minimum reserve capacity as a hard constraint and economic cost benchmark as the objective, integrates a multi-dimensional dataset across the entire domain to construct a hierarchical multi-objective optimization model, and solves for the optimal decision scheme. The personalized adjustment module, based on resource characteristic data in the multi-dimensional data of the whole domain, decomposes the optimal decision scheme into personalized instruction outputs that adapt to each resource. Each regional scheduling node receives the personalized instructions and executes the processing adjustment instructions.

2. The AGC coordinated optimization control system based on multi-source data fusion according to claim 1, characterized in that, The data-driven module filters out core data that are strongly related to risk assessment, working condition identification, and cost calculation from the multi-dimensional dataset, specifically including risk association dataset, working condition identification dataset, and cost calculation dataset. The risk-related dataset includes unit tripping power, sudden drop in renewable energy, load mutation, renewable energy penetration rate, unit health, and tie line transmission margin. The operating condition discrimination dataset includes real-time frequency, rated frequency, real-time load, and maximum load; The cost calculation dataset includes conventional unit resource types, energy storage resource types, market resource types, minimum output limits for each type of resource, maximum output limits for each type of resource, and unit regulation cost parameters.

3. The AGC coordinated optimization control system based on multi-source data fusion according to claim 2, characterized in that, In the data-driven module, the risk model expression is: ； in, This represents the output of the risk model, namely the minimum reserve capacity requirement; Represents the core items of risk quantification. The power deficit random variables represent the power of unit tripping, the magnitude of the sudden drop in renewable energy, and the load change, corresponding to each single disturbance scenario. Represents a risk quantification coefficient; Represents the risk deviation correction term; The weighting coefficient represents the penetration rate of new energy sources; Represents the penetration rate of new energy sources; Represents the weighting coefficient of unit health; Represents the health status of the generator set; Represents the transmission margin weighting coefficient for tie lines; This represents the transmission margin of the tie line.

4. The AGC coordinated optimization control system based on multi-source data fusion according to claim 2, characterized in that, The data-driven module calculates frequency deviation and load rate based on the operating condition discrimination dataset to determine the operating condition type. The calculation formulas for frequency deviation and load rate are as follows: ； ； in, Represents frequency deviation; Represents real-time frequency; Represents the rated frequency; Represents load factor; Represents real-time load; This represents the maximum load.

5. The AGC coordinated optimization control system based on multi-source data fusion according to claim 4, characterized in that, The method for determining the type of operating condition is as follows: when , representing the disturbance condition; when and This represents peak operating conditions; when and This represents a low-end operating condition; when and , representing a flat operating condition.

6. The AGC coordinated optimization control system based on multi-source data fusion according to claim 5, characterized in that, In the data-driven module, the economic cost benchmark formula expression for different operating conditions is as follows: ； in, Representing the Economic cost benchmark for similar operating conditions; Represents the type of resource to be adjusted. Represents conventional generating units. Represents energy storage. Represents market resources; Representing the Class resources in the Unit adjustment cost under similar operating conditions; This represents the maximum acceptable unit cost of backup. Represents an indicator function; This means that only under disturbed or peak operating conditions, the cost of additional backup redundancy is added. This represents the cost of backup redundancy.

7. The AGC coordinated optimization control system based on multi-source data fusion according to claim 6, characterized in that, In the intelligent optimization module, the hierarchical multi-objective optimization model is expressed by the following formula: ； ； in, This represents the objective function to be optimized; Representing the Within the scheduling period, the first Class resources in the The optimal decision-making scheme under similar working conditions, i.e., the output result of the hierarchical multi-objective optimization model; Representing the Total system reserve capacity during each scheduling period; , Represents weight; Represents constraints; This represents a core security constraint, mandating that the total system backup capacity must not be less than the minimum backup capacity; This means that the optimal capacity allocation is limited between the minimum and maximum output limits. Representing the Class resources in the Minimum output limit under similar working conditions Representing the Class resources in the Maximum output limit under similar working conditions.

8. The AGC coordinated optimization control system based on multi-source data fusion according to claim 7, characterized in that, The The calculation formula is: 。 9. The AGC coordinated optimization control system based on multi-source data fusion according to claim 8, characterized in that, The personalized adjustment module extracts all resources from the full-domain multi-dimensional data set. The characteristic parameters are categorized by resource type: conventional units Minimum start-up and shutdown time, ramp rate, coal consumption curve, and peak shaving range; Energy storage Charge / discharge efficiency, capacity limit, response time delay, and cycle life constraints; Market Resources Price quote curve, call response time, and adjustable capacity range; Match the working condition type of the current scheduling period Retrieve the threshold values ​​of resource characteristic parameters under the corresponding operating conditions.

10. The AGC coordinated optimization control system based on multi-source data fusion according to claim 9, characterized in that, The personalized adjustment module will output the hierarchical multi-objective optimization model. By splitting by resource type, the target allocated capacity of a single resource under the current operating conditions is obtained and marked as follows. Then to Perform hard constraint verification; if the physical limits are exceeded, correct according to the rules: like ,but The difference is allocated to other resources according to the principle of optimal cost. like ,but The excess portion is included in the system's backup redundancy; like ,but ; Representing the Within the scheduling period, the first Class resources in the Final executable allocation capacity under similar operating conditions; The revised Convert the command parameters into executable parameters for each resource: Conventional generating units: converted to output adjustment commands; Energy storage: converted to charge / discharge mode commands; Market resources: converted to capacity call commands.

Images