Power grid operation scene generation method and system, electronic equipment and storage medium

Abstract

The invention discloses a power grid operation scene generation method and system, electronic equipment and a storage medium. The method comprises the steps that scene generation conditions are set; analyzing the scene generation condition, and generating a network topology; according to scene generation conditions, an anticipated fault set adapting to different scene generation requirements is constructed, and network topology is updated; obtaining operation states of a new energy station and a load, and generating wind-light load time sequence data according with a wind-light power generation rule and a load power utilization rule of a corresponding region and a corresponding time period in combination with a scene generation condition and a network topology adapted to a current scene generation demand; performing unit combination optimization based on the wind-light load time sequence data, and obtaining unit combination and active power under the scene generation condition; generating power flow data based on the unit combination and the active power under the scene generation condition; and carrying out association combination on the power flow data and the environment data to generate power grid operation scene data. According to the invention, the power grid operation scene conforming to the physical law and the operation constraint can be automatically generated according to the service demand.

Description

Technical Field

[0001] This invention belongs to the field of power system technology, specifically relating to a method, system, electronic device, and storage medium for generating power grid operation scenarios. Background Technology

[0002] In traditional dispatching models, dispatchers arrange grid operation modes based on factors such as load demand, power generation capacity, grid transmission capacity, equipment maintenance needs, and reserve requirements. This generates data on operation modes and capacity limits, unit combinations, output curves, tie-line plans, and maintenance plans. This approach has limited predictable conditions, and mode arrangement is influenced by human experience. With the rapid development of new power systems, renewable energy output and load power are significantly affected by weather conditions, especially under extreme weather conditions. The number of different power system operation modes has surged. Verification of dispatch automation technology, training of artificial intelligence applications, grid situation simulation, and identification of grid safety boundaries all require multi-scenario verification of grid operation to cope with the random changes in new power systems.

[0003] The essence of generating power grid operation scenarios is to set some known conditions and obtain a set of data describing the power grid's operating state. Based on the scenario definition and description, a set of internal and external power grid operation data is generated, quantitatively describing the internal and external environment of the power grid, equipment status, and dynamic trends of power flow changes. The power grid is a massive, real-time balanced system where power generation and consumption must be constantly balanced, and its state is constantly changing. Influenced by external environment, electricity consumption behavior, power grid planning, equipment status, and other factors, the power system's generation, transmission, transformation, distribution, and consumption are affected by different factors that are mutually coupled, making it difficult to use a single method to globally generate power grid operation scenario data.

[0004] Currently, in the patent applications and authorizations for inventions in the power system field, scenario generation technologies are typically decoupled, including the generation of new energy power generation scenarios on the source side (such as publication numbers CN120849990A and CN120851298A) and the generation of power load scenarios on the load side (such as publication numbers CN120896146A and CN120317761A). These existing technologies consider scenarios corresponding to the uncertain fluctuations of wind, solar, and load, but often focus on the description of local domains, lacking a holistic reflection of the overall power grid operation status and failing to provide a global perspective on the power grid's operation. Some technologies also attempt to consider the interaction between sources and loads and the operational coupling characteristics between them to obtain uncertain scenarios (such as CN120911792A and CN120807223A), but because they fail to fully consider the interactions between all key factors, they are still insufficient to comprehensively and accurately simulate changes in the overall power grid operation status. More importantly, existing technologies cannot proactively construct a physically consistent complete operation scenario that strictly matches the user-specified composite business conditions (such as equipment failure caused by extreme weather).

[0005] Therefore, given the high proportion of renewable energy integration in new power systems and the frequent occurrence of extreme events, there is an urgent need for a technical solution that can automatically, collaboratively, and completely generate a power grid operation scenario that includes all elements such as topology, timing, faults, and power flow based on specified business conditions. Currently, there is no publicly available technology that can achieve this goal. Summary of the Invention

[0006] The purpose of this invention is to address the problems in the prior art by providing a method, system, electronic device, and storage medium for generating power grid operation scenarios. This invention can automatically generate power grid operation scenarios that conform to physical laws and operational constraints based on business needs. Based on the inherent mechanism of power system operation, it incorporates key elements such as meteorological conditions, disaster events, primary energy supply, equipment availability status (including operation, standby, maintenance, and fault probability), load demand and regulation capacity, output characteristics of new energy sources (centralized or distributed), and power generation capacity of conventional units into a unified generation framework. Under the premise of meeting the dispatch plan boundary and safety and stability constraints, it collaboratively generates a complete operation scenario that includes network topology, time-series power, fault prediction set, and power flow distribution, ensuring the physical consistency, business rationality, and multi-scenario coverage of the scenario.

[0007] To achieve the above objectives, the present invention provides the following technical solution: Firstly, a method for generating power grid operation scenarios is provided, including: Set the scene generation conditions; Analyze the scenario generation conditions and generate the network topology; Based on the scenario generation conditions, construct a set of anticipated faults that adapt to the generation requirements of different scenarios, and update the network topology; The system acquires the operating status of new energy power plants and loads, and combines the scenario generation conditions with the network topology that adapts to the current scenario generation requirements to generate wind and solar load time-series data that conforms to the wind and solar power generation patterns and load electricity consumption patterns of the corresponding region and time period. Based on wind, solar and load time series data, unit combination optimization is performed to obtain unit combination and active power under the generated scenario conditions; Power flow data is generated based on the unit combination and active power under the scenario generation conditions. By linking and combining power flow data and environmental data, power grid operation scenario data can be generated.

[0008] As a preferred embodiment, the step of setting scenario generation conditions includes any one or more combinations of time and location, operating indicators, power grid planning, primary energy sources, disaster events, meteorological environment, dispatching plans, and load regulation capabilities. Among them, time and location are mandatory, while other conditions are optional; The time and location settings specify the time period, data sampling interval, and power grid range. Based on the specified time, the system extracts the matching power grid model version and power grid operation data. For historical moments, it retrieves the corresponding historical power grid data, and for future moments, it retrieves the corresponding planned forecast data. If historical data or planned forecast data is missing, it combines other settings to retrieve similar daily power grid operation data. When only the time and location conditions are specified, the system reads the specified power grid model and the operation data for the specified time period through the scenario data service. The operational indicators are set as target values ​​for the generated power grid operation scenarios. These indicators include categories such as safety and stability, balance and regulation, renewable energy consumption, and power grid overview. Among them, the safety and stability category sets static safety indicators, the balance and regulation category includes balance capacity and regulation capacity, the renewable energy consumption category is renewable energy obstruction indicators, and the power grid overview indicators include AC / DC channel disaster values ​​and levels, as well as the scale of equipment failures under disasters. Based on the operational indicators, similar daily power grid models and operational data are obtained through indicator label matching. With the goal of minimizing the adjustment amount, the indicators are converted into power grid generation and consumption power and network channel power. The power grid planning settings include new and expanded power plants, substations, new energy power stations, distributed photovoltaics, AC / DC lines, transformers, various reactive power compensations, and changes on all sides of the power grid, including generation, transmission, transformation, distribution, and consumption. Based on the settings, the system checks whether the power grid planning equipment has been modeled in the existing model version. If it has been modeled, it can be used directly; if it has not been modeled, a new model should be created in the future state model. Primary energy settings include coal storage, gas storage, and water inflow conditions, which convert relevant data into the total electricity that the power plant can generate. The disaster event settings include the disaster event type and the time and location of the event; based on the disaster event, using the anticipated faults that take into account external disasters, the disaster event is transformed into a change in the power grid operation status. On the initial power grid operation data extracted or generated by the scenario, faults and power fluctuations are superimposed to generate power grid operation mode data under the disaster scenario; The meteorological environment settings include meteorological type settings. The meteorological environment time series data and time and location settings are matched to conform to the climate patterns of the external environment of the power grid at the specified time and region. Based on the set meteorological environment time series data, the matching daily meteorological data and operation mode data of the power grid are obtained, and combined with the source and load time series data to generate and transform the set conditions into active power time series data of new energy, distributed photovoltaic and load. The dispatch plan sets price factors, reserve constraints, medium- and long-term transactions, spot transactions, maintenance plans, and tie-line plan conditions. It determines whether the specified time period is a historical or future time. If it is a historical time, it generates active power generation data and tie-line exchange power data based on historical planned forecasts and actual operating data, according to the settings and in conjunction with the dispatch plan function. If it is a future time and planned forecast data is available, it adjusts the data based on the planned forecast data, according to the set conditions, and in conjunction with the dispatch plan function, and extrapolates future trends. If it is a future time but planned forecast data is missing, it generates planned forecast data using the dispatch plan function and extrapolates future trends. The load regulation capacity is set according to the power supply area of ​​equivalent load, including the load scale and its adjustable or interruptible capacity of industrial users, the load scale and its adjustable capacity of commercial and public building temperature control and lighting, the load scale and its adjustable capacity of residential users, the load scale and its adjustable capacity of electric vehicles, and the scale and its adjustable capacity of aggregators. Based on the conditions, the adjustable capacity is quantified to generate load response speed, regulation power and capacity, response direction and control method.

[0009] As a preferred approach, the method also includes a conflict verification step for the set scenario generation conditions. The conflict verification includes the degree of matching between time, location and meteorological environment, disaster events, dispatch plans, power grid planning, primary energy, operating indicators and load regulation capabilities. Based on the analysis and statistics of the qualitative or quantitative correlation between various scenario generation conditions, verification rules are formed.

[0010] As a preferred approach, in the step of generating network topology under the analysis scenario generation conditions, the power grid model and the on / off status or switching status of the equipment are obtained, and topology analysis is performed to generate the power grid network topology and corresponding calculation model under the scenario generation conditions.

[0011] As a preferred approach, in the step of constructing a set of anticipated faults to meet the generation requirements of different scenarios based on the scenario generation conditions and updating the network topology, preset external meteorological conditions, equipment parameters, meteorological data during historical similar fault periods, and power grid operation data are obtained. A probability model is established for single equipment faults, a set of clustered combined faults is generated, and the IDs of all faulted equipment in the fault set under the corresponding meteorological conditions are output.

[0012] As a preferred approach, in the step of generating a set of clustered combined faults, the probability of power grid equipment failure caused by the external environment within a certain period of time is dynamically assessed based on the security and stability assessment requirements of different business applications, combined with the external environment, geographical location, and equipment design parameters, to generate a set of anticipated faults that adapt to different scenarios; the propagation and evolution process of a single fault leading to a chain of faults is analyzed, the types of chain faults and their occurrence scenarios are screened out, and a set of anticipated power grid clustered faults within a certain period of time is generated; the generation of the set of anticipated faults considering natural environmental factors starts from the mechanism of natural disasters affecting equipment, establishes a model of the probability of equipment failure caused by natural disasters, and then calculates the probability of clustered anticipated faults based on the assessment of the probability of single equipment failure caused by disasters, thereby generating the set of anticipated faults.

[0013] As a preferred approach, when acquiring the operating status of new energy power plants and loads, and combining the scenario generation conditions and network topology adapted to the current scenario generation requirements to generate wind and solar load time-series data that conforms to the wind and solar power generation patterns and load consumption patterns of the corresponding region and time period, a set of deterministic expected scenarios or multiple sets of random scenario source-load time-series data within the feasible domain are generated according to the scenario generation conditions. One set of time-series data includes centralized wind and solar power output data and distributed solar power output data, as well as load forecast data, within the corresponding region, based on the number of new energy power plants, equivalent loads, and equivalent generating units. Several sets of data for wind farms, photovoltaic stations, distributed photovoltaics, and bus loads are generated, along with the same centralized wind and solar power output data, distributed photovoltaic data, and load data. Multiple sets of data for new energy stations, equivalent loads, and equivalent generating units can be decomposed. Multiple sets of time-series data include time-series data within the feasible domain that conform to the fluctuation patterns of wind power, photovoltaics, distributed photovoltaics, and loads under specified conditions. The generated wind, solar, and load time-series data conform to the inherent correlation patterns of various factors and wind and solar power output and load electricity consumption within the specified region and time period, and conform to the fluctuation patterns of wind and solar power output and load within the specified region and time period.

[0014] As a preferred embodiment, the minimum time unit for the wind and solar load time series data is 1 hour, and the minimum time interval is 1 minute. Time series data for 1 to 24 hours is generated at a time interval of 1, 5, 15, or 60 minutes, supporting the generation of time series data with different durations and time intervals. For data with a week's duration, the time interval is no more than 15 minutes; for data with a month's or year's duration, the time interval is 60 minutes. When time series data for the next day needs to be generated, it is generated on a daily basis, using the results generated that day as input for the data generated the next day, thus generating the data in a rolling manner. When specifying the wind and solar power generation or load power consumption within a certain period, the generated time series data meets the power consumption constraints.

[0015] As a preferred approach, when optimizing the unit combination based on wind, solar and load time series data to obtain the unit combination and active power under the scenario generation conditions, the network topology, wind, solar and load time series data, port plan, tie line power of similar days, and conventional unit power generation plan of the day to be generated are obtained, the unit combination optimization problem is solved, and the unit start-up and shutdown mode and active power data under the specified scenario generation conditions are output. In the steps of solving the unit combination optimization problem, the provincial power grid is decomposed into sub-plans, and the total power is distributed to each line using the historical proportion method based on similar days or the manually specified proportion method. The method of allocating total power to each line based on the historical proportion of similar days includes the following steps: For each similar day, the total active power transmitted by each tie line member within the same time period is counted as a percentage of the tie line group. If multiple similar days exist, the total planned tie line power for the day to be generated is allocated proportionally based on time proximity. The percentage of the measured power of the i-th tie line member on the d-th similar day to the total planned power of the tie line is calculated using the following formula:

[0016] The weighted recommendation percentage is calculated using the following expression:

[0017] Total planned power of the generated day The proportional allocation is calculated as follows:

[0018] The manual allocation ratio method involves manually setting a fixed allocation coefficient for each tie line group and tie line member. The implementation methods include the following: Method 1: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios; for each set of source-load random scenarios, independently call the complete power generation plan optimization model to generate unit combination and output data, and calculate the average of multiple sets of unit combination and output data to obtain unit power generation time series data. Method 2: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios. For each set of source-load random scenarios, obtain the corresponding equivalent load time-series active power data. Use a clustering analysis algorithm to cluster the multiple sets of data into groups. Take the median equivalent load time-series data of all data in the group and call the power generation planning model to generate unit combinations. Then, superimpose all the original new energy power generation time-series data and load active power data in the corresponding group, and adjust the unit output in combination with AC power flow to obtain the unit time-series output data for each source-load scenario. The processing method is the same for each group. Method 3: Based on the specified time and location, operating indicators, power grid planning, primary energy sources, disaster events, or meteorological conditions, generate the most probable set of source-load operating scenarios; call the power generation planning model to obtain power generation planning data, set optimization objectives, and call the Safety Constraint Unit Combination (SCUC) to generate unit combinations; on this basis, superimpose multiple sets of unconditional random combination data to adjust unit output; Method 4: Using artificial intelligence, multiple sets of unconditional source-load random scenarios are generated by specifying only the time and region, and the power generation planning model is called. For each set of source-load data, a corresponding power generation plan is obtained, and the unit combination and active power output are solved through a large model.

[0019] As a preferred embodiment, in the step of generating power flow data based on the unit combination and active power under the scenario generation conditions, a power flow dataset is generated by acquiring the network topology, wind-solar-load time-series data, unit combination and active power, similar day grid model and operation data under the scenario generation conditions of the day to be generated; the power flow data includes the voltage amplitude, phase angle, branch power and reactive power of each node.

[0020] As a preferred approach, this also includes verifying the rationality of the cross-sections and time-series data at each moment after generating power grid operation scenario data, selecting appropriate indicators to classify the scenario data according to business needs, and eliminating redundant scenarios.

[0021] Secondly, a power grid operation scenario generation system is provided, including: The condition setting module is used to set the scene generation conditions; The network topology generation module is used to analyze the scene generation conditions and generate the network topology. The contingency set construction module is used to construct contingency sets that adapt to different scenario generation requirements based on scenario generation conditions, and update the network topology; The wind-solar-load time-series data generation module is used to obtain the operating status of new energy power plants and loads, and combine the scene generation conditions and the network topology that adapts to the current scene generation requirements to generate wind-solar-load time-series data that conforms to the wind and solar power generation patterns and load electricity consumption patterns of the corresponding region and time period. The power generation and transmission time series data generation module is used to optimize the unit combination based on wind, solar and load time series data, and obtain the unit combination and active power under the scenario generation conditions. The power flow data generation module is used to generate power flow data based on the unit combination and active power under the scenario generation conditions. The scenario data generation module is used to associate and combine power flow data and environmental data to generate power grid operation scenario data.

[0022] As a preferred embodiment, the scenario generation conditions set by the condition setting module include any one or more combinations of time and location, operating indicators, power grid planning, primary energy, disaster events, meteorological environment, dispatching plans, and load regulation capabilities; Among them, time and location are mandatory, while other conditions are optional; The time and location settings specify the time period, data sampling interval, and power grid range. Based on the specified time, the system extracts the matching power grid model version and power grid operation data. For historical moments, it retrieves the corresponding historical power grid data, and for future moments, it retrieves the corresponding planned forecast data. If historical data or planned forecast data is missing, it combines other settings to retrieve similar daily power grid operation data. When only the time and location conditions are specified, the system reads the specified power grid model and the operation data for the specified time period through the scenario data service. The operational indicators are set as target values ​​for the generated power grid operation scenarios. These indicators include categories such as safety and stability, balance and regulation, renewable energy consumption, and power grid overview. Among them, the safety and stability category sets static safety indicators, the balance and regulation category includes balance capacity and regulation capacity, the renewable energy consumption category is renewable energy obstruction indicators, and the power grid overview indicators include AC / DC channel disaster values ​​and levels, as well as the scale of equipment failures under disasters. Based on the operational indicators, similar daily power grid models and operational data are obtained through indicator label matching. With the goal of minimizing the adjustment amount, the indicators are converted into power grid generation and consumption power and network channel power. The power grid planning settings include new and expanded power plants, substations, new energy power stations, distributed photovoltaics, AC / DC lines, transformers, various reactive power compensations, and changes on all sides of the power grid, including generation, transmission, transformation, distribution, and consumption. Based on the settings, the system checks whether the power grid planning equipment has been modeled in the existing model version. If it has been modeled, it can be used directly; if it has not been modeled, a new model should be created in the future state model. Primary energy settings include coal storage, gas storage, and water inflow conditions, which convert relevant data into the total electricity that the power plant can generate. The disaster event settings include the disaster event type and the time and location of the event; based on the disaster event, using the anticipated faults that take into account external disasters, the disaster event is transformed into a change in the power grid operation status. On the initial power grid operation data extracted or generated by the scenario, faults and power fluctuations are superimposed to generate power grid operation mode data under the disaster scenario; The meteorological environment settings include meteorological type settings. The meteorological environment time series data and time and location settings are matched to conform to the climate patterns of the external environment of the power grid at the specified time and region. Based on the set meteorological environment time series data, the matching daily meteorological data and operation mode data of the power grid are obtained, and combined with the source and load time series data to generate and transform the set conditions into active power time series data of new energy, distributed photovoltaic and load. The dispatch plan sets price factors, reserve constraints, medium- and long-term transactions, spot transactions, maintenance plans, and tie-line plan conditions. It determines whether the specified time period is a historical or future time. If it is a historical time, it generates active power generation data and tie-line exchange power data based on historical planned forecasts and actual operating data, according to the settings and in conjunction with the dispatch plan function. If it is a future time and planned forecast data is available, it adjusts the data based on the planned forecast data, according to the set conditions, and in conjunction with the dispatch plan function, and extrapolates future trends. If it is a future time but planned forecast data is missing, it generates planned forecast data using the dispatch plan function and extrapolates future trends. The load regulation capacity is set according to the power supply area of ​​equivalent load, including the load scale and its adjustable or interruptible capacity of industrial users, the load scale and its adjustable capacity of commercial and public building temperature control and lighting, the load scale and its adjustable capacity of residential users, the load scale and its adjustable capacity of electric vehicles, and the scale and its adjustable capacity of aggregators. Based on the conditions, the adjustable capacity is quantified to generate load response speed, regulation power and capacity, response direction and control method.

[0023] As a preferred option, a condition verification module is also included, which is used to perform conflict verification on the set scenario generation conditions. Conflict verification includes the degree of matching between time, location and meteorological environment, disaster events, scheduling plans, power grid planning, primary energy, operating indicators and load regulation capabilities. Based on the analysis and statistics of the qualitative or quantitative correlation between various scenario generation conditions, verification rules are formed.

[0024] As a preferred embodiment, the network topology generation module obtains the power grid model and the on / off status or switching status of the equipment, and performs topology analysis to generate the power grid network topology and corresponding calculation model under the scenario generation conditions.

[0025] As a preferred embodiment, the anticipated fault set construction module acquires preset external meteorological conditions, equipment parameters, meteorological data during historical similar fault periods, and power grid operation data, establishes a probability model for single equipment faults, generates a clustered combined fault set, and outputs the IDs of all faulty equipment in the fault set under the corresponding meteorological conditions.

[0026] As a preferred embodiment, the anticipated fault set construction module, when generating a clustered combined fault set, dynamically assesses the probability of power grid equipment failure caused by the external environment over a period of time, taking into account the security and stability assessment needs of different business applications, and combining external environment, geographical location, and equipment design parameters, to generate anticipated fault sets adapted to different scenarios; it analyzes the propagation and evolution process of cascading failures caused by a single fault, filters out cascading failure types and occurrence scenarios, and generates a clustered fault set for the power grid over a period of time; considering the generation of anticipated fault sets based on natural environmental factors, it starts from the mechanism of natural disasters affecting equipment, establishes a model of the probability of equipment failure caused by natural disasters, and then calculates the probability of clustered anticipated faults based on the assessment of the probability of single equipment failure caused by disasters, generating the anticipated fault set.

[0027] As a preferred embodiment, the wind-solar-load time-series data generation module generates a set of deterministic expected scenarios or multiple sets of random scenario source-load time-series data within the feasible domain, based on scenario generation conditions. The set of time-series data includes centralized wind power and photovoltaic (PV) and distributed PV output data, load forecast data, and generates several data entries for wind farms, PV stations, distributed PV, and bus loads based on the number of new energy power plants, equivalent loads, and equivalent generating units. It also includes identical centralized wind-solar output data, distributed PV data, and load data, and can be decomposed into multiple sets of new energy power plants, equivalent loads, and equivalent generating unit data. Multiple sets of time-series data include multiple sets of feasible domain time-series data that conform to the fluctuation patterns of wind power, PV, distributed PV, and load within specified conditions. The generated wind-solar-load time-series data conforms to the inherent correlation patterns of various factors and wind-solar output and load electricity consumption within the specified region and time period, and conforms to the fluctuation patterns of wind-solar output and load within the specified region and time period.

[0028] As a preferred embodiment, the wind-solar-load time-series data generation module generates wind-solar-load time-series data with a minimum time length of 1 hour and a minimum time interval of 1 minute. It generates time-series data for 1 to 24 hours at a time, with time intervals of 1, 5, 15, or 60 minutes, supporting the generation of time-series data with different durations and time intervals. For data with a week's duration, the time interval is no greater than 15 minutes; for data with a month's or year's duration, the time interval is 60 minutes. When time-series data for the next day needs to be generated, it is generated on a daily basis, using the results generated that day as input for the data generated the next day, thus generating data in a rolling manner. When specifying the wind and solar power generation or load power consumption within a certain period, the generated time-series data meets the power consumption constraints.

[0029] As a preferred embodiment, the power generation and transmission time series data generation module acquires the network topology, wind and solar load time series data, grid connection plan, tie line power of similar days, and conventional unit power generation plan for the day to be generated, solves the unit combination optimization problem, and outputs the unit start-up and shutdown mode and active power data under the specified scenario generation conditions. When solving the unit combination optimization problem, the provincial power grid is decomposed into sub-plans, and the total power is distributed to each line using the historical ratio method based on similar days or the manually specified ratio method. The method of allocating total power to each line based on the historical proportion of similar days includes the following steps: For each similar day, the total active power transmitted by each tie line member within the same time period is counted as a percentage of the tie line group. If multiple similar days exist, the total planned tie line power for the day to be generated is allocated proportionally based on time proximity. The percentage of the measured power of the i-th tie line member on the d-th similar day to the total planned power of the tie line is calculated using the following formula:

[0030] The weighted recommendation percentage is calculated using the following expression:

[0031] Total planned power of the generated day The proportional allocation is calculated as follows:

[0032] The manual allocation ratio method involves manually setting a fixed allocation coefficient for each tie line group and tie line member. The implementation methods include the following: Method 1: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios; for each set of source-load random scenarios, independently call the complete power generation plan optimization model to generate unit combination and output data, and calculate the average of multiple sets of unit combination and output data to obtain unit power generation time series data. Method 2: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios. For each set of source-load random scenarios, obtain the corresponding equivalent load time-series active power data. Use a clustering analysis algorithm to cluster the multiple sets of data into groups. Take the median equivalent load time-series data of all data in the group and call the power generation planning model to generate unit combinations. Then, superimpose all the original new energy power generation time-series data and load active power data in the corresponding group, and adjust the unit output in combination with AC power flow to obtain the unit time-series output data for each source-load scenario. The processing method is the same for each group. Method 3: Based on the specified time and location, operating indicators, power grid planning, primary energy sources, disaster events, or meteorological conditions, generate the most probable set of source-load operating scenarios; call the power generation planning model to obtain power generation planning data, set optimization objectives, and call the Safety Constraint Unit Combination (SCUC) to generate unit combinations; on this basis, superimpose multiple sets of unconditional random combination data to adjust unit output; Method 4: Using artificial intelligence, multiple sets of unconditional source-load random scenarios are generated by specifying only the time and region, and the power generation planning model is called. For each set of source-load data, a corresponding power generation plan is obtained, and the unit combination and active power output are solved through a large model.

[0033] As a preferred embodiment, the power flow data generation module generates a power flow dataset by acquiring network topology, wind-solar-load time-series data, unit combination and active power, similar day grid model and operation data under the scenario generation conditions of the day to be generated; the power flow data includes the voltage amplitude, phase angle, branch power and reactive power of each node.

[0034] As a preferred option, a scenario verification module is also included, which is used to verify the rationality of the cross section and time series data at each moment after generating power grid operation scenario data, select appropriate indicators to classify the scenario data according to business needs, and eliminate redundant scenarios.

[0035] Thirdly, an electronic device is provided, including a processor and a memory, the processor being configured to execute a computer program stored in the memory to implement the power grid operation scenario generation method as described in the first aspect.

[0036] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing at least one instruction, which, when executed by a processor, implements the power grid operation scenario generation method as described in the first aspect.

[0037] Compared with the prior art, the first aspect of the present invention has at least the following beneficial effects: The power grid operation scenario generation method proposed in this invention can set scenario generation conditions (including time and location, operation indicators, power grid planning and operation, primary energy supply, disaster events, meteorological conditions, dispatch plan boundaries, etc.) according to business needs. It can also acquire initial basic data (including meteorological information, disaster information, primary energy supply, power grid planning, power grid model and operation data, geographic information, etc.) based on these conditions, and generate power grid operation scenarios under specified conditions. This method can comprehensively reflect the overall operation status of the entire power grid, providing a global perspective on power grid operation. By acquiring the operation status of new energy power plants and loads, combined with scenario generation conditions and network topology adapted to the current scenario generation requirements, it generates wind-solar-load time-series data that conforms to the wind and solar power generation patterns and load consumption patterns of the corresponding region and time period. Simultaneously, based on the wind-solar-load time-series data, it optimizes unit combination, obtains unit combination and active power under the scenario generation conditions, and fully considers the interaction between all key factors to comprehensively and accurately simulate changes in the overall operation status of the power grid. This invention enables condition-driven generation of complete scenarios. Users can set any combination of business conditions (such as "high temperature on a certain day + main transformer maintenance + large-scale renewable energy generation"), and automatically construct a matching power grid operating state accordingly. This overcomes the limitations of traditional methods that rely on historical playback or random sampling, significantly improving the business fit and usability of the scenario. This invention ensures the physical consistency and system integrity of the scenario. By sequentially executing network topology generation, wind, solar, load, and power transmission time-series data generation, fault configuration, and power flow calculation, it ensures logical consistency and physical feasibility among equipment states, power injection, network structure, and power flow results in the generated scenario, avoiding infeasibility or contradictory states caused by partial data splicing. Furthermore, this invention supports the data requirements of intelligent applications. The output structured and labeled scenarios can be directly used for simulation, AI model training, risk assessment, and other scenarios, and can be incorporated into a big data platform for asset management, thereby providing a reliable data foundation for the digital operation of new power systems.

[0038] It is understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 A schematic diagram of the overall architecture of the power grid operation scenario generation system according to an embodiment of the present invention; Figure 2 Overall flowchart of the power grid operation scenario generation method according to an embodiment of the present invention; Figure 3 A flowchart illustrating the generation of network topology in this embodiment of the invention; Figure 4 A flowchart illustrating the construction of a set of anticipated faults according to an embodiment of the present invention; Figure 5 A flowchart illustrating the generation of wind-solar-load time-series data in an embodiment of the present invention; Figure 6 A flowchart illustrating the generation and transmission timing data in this embodiment of the invention; Figure 7 A flowchart illustrating the generation of power flow data in an embodiment of the present invention. Detailed Implementation

[0041] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail. Flowcharts are used in the embodiments of this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more steps may be removed from these processes.

[0042] To address the problem that existing power grid operation scenario generation methods struggle to automatically generate complete and physically consistent power grid operation states based on user-specified multi-dimensional business conditions (such as time and location, weather events, equipment maintenance plans, and dispatch boundaries), this invention provides a power grid operation scenario generation method that can adapt to the needs of new power systems. It aims to support advanced applications such as dispatch automation verification, artificial intelligence training, security boundary identification, and situational simulation, thereby enhancing the power grid's adaptability and resilience to high-proportion renewable energy integration and extreme event impacts.

[0043] Please see Figure 2 The method for generating power grid operation scenarios according to embodiments of the present invention mainly includes the following steps: S1. Set scene generation conditions; S2. Analyze the scene generation conditions and generate the network topology; S3. Based on the scenario generation conditions, construct a set of anticipated faults that adapt to the generation requirements of different scenarios, and update the network topology; S4. Obtain the operating status of new energy power plants and loads, and combine the scenario generation conditions and network topology that are adapted to the current scenario generation requirements to generate wind and solar load time-series data that conforms to the wind and solar power generation patterns and load electricity consumption patterns of the corresponding region and time period. S5. Optimize the unit combination based on wind, solar and load time series data to obtain the unit combination and active power under the scenario generation conditions; S6. Generate power flow data based on the unit combination and active power under the scenario generation conditions; S7. Combine and correlate power flow data and environmental data to generate power grid operation scenario data.

[0044] In one possible implementation, step S1 obtains all the necessary basic data in batches through the scenario data service, providing scenario setting conditions and basic data inside and outside the power grid for subsequent functions.

[0045] The scenario generation conditions include any one or more combinations of time and location, operating indicators, power grid planning, primary energy, disaster events, meteorological environment, dispatching plans, and load regulation capacity; among them, time and location are mandatory, while other conditions are optional.

[0046] The time and location settings specify the time period, data sampling interval, and power grid range. This condition serves as the basis for extracting the power grid model and initial operating data. Based on the specified time, it extracts the matching power grid model version and power grid operating data. For historical moments, it retrieves corresponding historical power grid data; for future moments, it retrieves corresponding planned forecast data. If historical or planned forecast data is missing, it combines other settings to obtain power grid operating data for similar days. When only this condition is specified, the operating data of the specified power grid model and time period is read through the scenario data service.

[0047] The operation indicator setting mainly involves setting target values ​​for indicators in the generated power grid operation scenarios. Specific indicators include those related to safety and stability, balance and regulation, renewable energy consumption, and power grid overview. The safety and stability category primarily sets static safety indicators. Balance and regulation indicators mainly include balancing capacity (spinning reserve, negative reserve, dispatchable load, ultra-short-term balance margin, etc.) and regulation capacity (primary frequency regulation capacity, AGC regulation capacity, voltage regulation capacity). Renewable energy consumption indicators mainly include renewable energy obstruction indicators. Power grid overview indicators include AC / DC channel disaster values ​​and levels, and equipment failure scale under disasters. Based on the specified operation indicators, the system uses indicator tag matching from a big data platform to obtain similar daily power grid models and operation data. Utilizing optimization calculation functions such as optimal power flow, with the goal of minimizing adjustments, the indicators are converted into power generation and consumption and network channel power.

[0048] Power grid planning settings mainly include changes to the power grid's generation, transmission, distribution, and consumption sides, such as newly built or expanded power plants, substations, new energy power plants, distributed photovoltaic systems, AC / DC lines, transformers, various reactive power compensation systems, substation upgrades, and load growth. Based on these settings, the future-state modeling function is used to first check whether the power grid planning equipment has been modeled in existing model versions. Modeled equipment is used directly; for unmodeled equipment, new models are created in the future-state modeling. New equipment parameters use design parameters or typical parameters, and the newly created models are incorporated into the power grid model management system of the big data platform.

[0049] Primary energy settings mainly include conditions such as coal storage, gas storage, and water supply. Based on these settings, the relevant data is converted into the total amount of electricity that the power plant can generate.

[0050] The disaster event settings mainly include lightning, wildfires, ice storms, typhoons, earthquakes, floods, and geological disasters, along with the time and location of the events. Based on the disaster events, the system utilizes the fault prediction function that takes into account external disasters to transform the disaster events into changes in the power grid's operational status, such as power grid equipment failures and variations in power generation and consumption. Fault and power fluctuation data are then overlaid on the initial power grid operational data extracted from a big data platform or generated based on specific scenarios to ultimately generate power grid operational mode data under the disaster scenario.

[0051] Meteorological environment settings include the configuration of meteorological data such as temperature, humidity, solar radiation, rainfall, sandstorms, wind direction, and wind speed. The time-series meteorological data must match the time and location settings and conform to the climatic patterns of the external power grid environment at the specified time and region. For example, prolonged thunderstorms cannot be configured for the Xinjiang Uygur Autonomous Region's Shagohuang energy base, nor can extreme cold weather be configured for the Hainan power grid. During function development, it is necessary to match the time and location settings and provide corresponding constraints. Based on the configured meteorological data, similar daily meteorological data and operating mode data of the power grid are obtained. Combined with the source-load time-series data generation function (prediction, wind and solar power conversion calculation, etc.), the configured conditions are transformed into active power time-series data for new energy sources, distributed photovoltaics, and loads.

[0052] The dispatch plan settings include setting conditions such as price factors, reserve constraints, medium- and long-term transactions, spot transactions, maintenance plans, and tie-line plans. First, it determines whether the specified time period is historical or future. If it is historical, based on historical planned forecasts and actual operating data, it generates active power generation data and tie-line exchange power data according to the settings and in conjunction with the dispatch plan function. If it is future and planned forecast data is available, it adjusts the data based on the planned forecast data, according to the set conditions, and in conjunction with the dispatch plan function, and extrapolates future trends. If it is future but planned forecast data is missing, it generates planned forecast data using the dispatch plan function and extrapolates future trends.

[0053] Load regulation capacity settings allow for configuration of industrial user load scale and its adjustable and interruptible capacity, commercial and public building temperature control and lighting load scale and its adjustable capacity, residential user load scale and its adjustable capacity, electric vehicle load scale and its adjustable capacity, and aggregator scale and its adjustable capacity, based on the power supply area of ​​equivalent load. Based on the configured conditions, the adjustable capacity is quantified to generate various parameters for the load regulation model, including load response speed, regulation power and capacity, response direction, and control method (direct control or demand response, where demand response needs to consider uncertainty).

[0054] In one possible implementation, the present invention further includes a conflict verification step for the set scenario generation conditions. Conflict verification includes the degree of matching between time and location, meteorological environment, disaster events, dispatch plans, power grid planning, primary energy sources, operating indicators, and load regulation capabilities. Verification rules are formed by analyzing and statistically analyzing the qualitative or quantitative correlations between various conditions. On the one hand, the verification rules determine whether the condition settings are reasonable; on the other hand, based on the previously set conditions, the settable range of subsequently set conditions is limited to avoid unreasonable scenarios with conflicting conditions.

[0055] In one possible implementation, the network topology generation in step S2 mainly includes the normal operation mode and the power grid equipment commissioning and decommissioning status at various times considering the maintenance plan, and determines the units that must be turned on and must be turned off according to the stable operation regulations.

[0056] like Figure 3 As shown, by acquiring the power grid model and the on / off status or switch status of equipment, and using the topology analysis function, a power grid network topology and its calculation model under specified scenario conditions are generated. This embodiment of the invention supports multiple data sources, and the appropriate data acquisition path can be selected according to the scenario configuration. If data is acquired from historical similar days, it is necessary to call the data service of a big data platform to obtain the power grid model and switch status for a certain day (or multiple candidate days). If stable operation regulations, mode arrangement expertise, and maintenance plans need to be considered, it is necessary to obtain the on / off status of equipment from the corresponding database and update the remote signaling values ​​of connected switches accordingly. Finally, based on the above power grid model and the finally determined switch status, topology analysis is performed to generate the final network topology.

[0057] In one possible implementation, step S3 involves acquiring preset external meteorological conditions, equipment parameters, meteorological data from historical similar fault periods, and power grid operation data. First, a probabilistic model is established for single-equipment faults. Then, a set of clustered combined faults is generated based on artificial intelligence methods. Finally, the IDs of all faulty equipment in the fault set under the given meteorological conditions are output, such as... Figure 4 As shown.

[0058] To address the safety and stability assessment needs of different business applications such as current power grid operation modes, planned operation modes, and dispatch operations, this system dynamically assesses the probability of power grid equipment failures caused by the external environment over a period of time by combining information such as external environment, geographical location, and equipment design parameters. It automatically generates a set of anticipated faults that adapt to different scenarios. By combining the characteristics of external environmental influences, power flow transfer, and expert experience, it analyzes the propagation and evolution process of cascading faults caused by a single fault, selects the key types of cascading faults and their occurrence scenarios, and automatically generates a set of anticipated power grid cluster faults over a period of time.

[0059] Intelligent generation of anticipated fault sets considering natural environmental factors typically starts from the mechanisms by which natural disasters such as typhoons, lightning, icing, and wildfires affect equipment. A model of the probability of equipment failure caused by natural disasters is established. Based on the assessment of the probability of individual equipment failure caused by the disaster, conditional probability or other assessment methods are used to calculate the probability of clustered anticipated faults, automatically generating a set of anticipated faults. Taking typhoons as an example, four pathways to equipment failure caused by typhoons are considered: foreign object snagging, wind-induced flashover, line breakage, and collapse. The probability of line failure is modeled and assessed, and the failure probability of the entire line is calculated based on these four probabilities. All lines are sorted according to their failure probabilities, and lines with failure probabilities greater than a set value are selected and included in the risk equipment set. In the risk equipment set, combinations are prioritized based on rules applicable to a certain disaster. If suitable combination rules cannot be extracted, an exhaustive combination and subsequent screening method is used to generate clustered combined faults.

[0060] In one possible implementation, when generating wind and solar load time-series data in step S4, active power time-series data of wind and solar load is generated based on the operating status of new energy power plants and loads, network topology, and external conditions such as meteorology and disasters.

[0061] Based on the set conditions, by acquiring gridded meteorological data, necessary equipment parameters, and historical wind and solar load time-series data, different algorithms are used to generate time-series active power data that conforms to the wind and solar power generation patterns and load consumption patterns of the region and time period. This includes time-series data for wind farms / clusters, photovoltaic stations / clusters, distributed photovoltaic areas, system loads, and bus loads at different voltage levels. Figure 5 As shown.

[0062] Step S4, when generating wind and solar load time series data, has the following functions: (1) It has the function of quantitative conversion of external environmental information such as meteorology, disasters, and primary energy supply, as well as the active power time-series evolution of source and load equipment status.

[0063] (2) It has the function of generating time-series running data of random fluctuation scenario of source load based on external information.

[0064] (3) It has the function of generating active power time sequence data of system, power supply analysis, substation and bus level source load, and the generated power data of different levels are coupled with each other and meet the constraints of total addition, network loss, ramping and congestion.

[0065] (4) It can generate a set of deterministic expected scenarios or multiple sets of random scenario source-load time series data within the feasible domain based on the specified conditions in the scenario condition settings or only the specified time and location (unconditional scenario generation). A set of time series data (deterministic expected scenarios, with the highest probability) includes: three sets of centralized wind power and photovoltaic, distributed photovoltaic and other new energy output data within the region, one set of system load forecast data, and several sets of wind farm, photovoltaic station, distributed photovoltaic and bus load data generated according to the number of new energy power stations, equivalent load and equivalent unit respectively. The same centralized wind and solar power output data, distributed photovoltaic data and load data can be decomposed into multiple sets of new energy power station, equivalent load and equivalent unit data; multiple sets of time series data (random scenarios within the feasible domain that meet the probability distribution) include: multiple sets of time series data within the feasible domain that meet the wind power, photovoltaic, distributed photovoltaic and load fluctuation rules within the specified conditions.

[0066] (5) The generated time series data must meet the following conditions: conform to the inherent correlation between meteorological factors and wind and solar power output and load power consumption in the specified area and time period; conform to the wind and solar power output and load fluctuation patterns in the specified area and time period, except for extreme scenarios of wind and solar loads specified separately; if it is a future period, it is necessary to comprehensively consider grid planning such as the expansion of the installed capacity of new energy and distributed power sources, and load growth rate and other factors.

[0067] (6) Load-side adjustment resources, as dispatchable quantities, are considered when generating the power grid operation mode. Currently, there is a lack of reported data on load adjustability; its adjustment capacity can be set according to demand.

[0068] (7) The minimum time unit for time series data is 1 hour, and the minimum time interval is 1 minute. A single batch can generate source-load time series active power data for 1 to 24 hours. The time interval can be selected from 1, 5, 15, and 60 minutes. The function should support the generation of time series data with different durations and time intervals. For data with a duration of one week, the time interval is recommended to be no more than 15 minutes. For data with a duration of one month or year, the time interval is recommended to be 60 minutes.

[0069] (8) When it is necessary to generate time series data for the next day, the results generated on the current day can be used as input for the data generated on the next day, and the source load time series data for the next day can be generated in a rolling manner.

[0070] (9) When specifying the power generation of wind and solar power or the power consumption of load within a certain period, the generated time series data should meet the power constraint requirements.

[0071] There are multiple technical approaches to generating wind, solar, and load time-series data. To address different business needs such as dispatch automation technology testing and verification, artificial intelligence training, power grid situation simulation, and safety and stability boundary analysis, a systematic approach should be taken, considering the quality and comprehensiveness of basic data, the completeness of scenario generation conditions, and the different business requirements for scenario data accuracy, generation speed, time intervals and durations, comprehensive coverage, randomness, and extreme scenarios. By comprehensively applying the above-mentioned different types of methods, a source-load scenario generation service system that meets various needs can be constructed. Combined with scenario generation intelligent agents, services can be combined on demand to generate power grid operation scenario data.

[0072] In one possible implementation, step S5 generates power generation and transmission time-series data. Based on wind, solar, and load active power time-series data and network topology, and combined with various constraints such as unit safety and economy, tie-line time-series data, unit combination, and output time-series data are generated. The generation and transmission time-series data generation includes power generation time-series data (i.e., active power time-series data of conventional thermal power units) and transmission time-series data (i.e., active power time-series data of tie-line). The active power output of the units is jointly determined by the unit generation plan and the tie-line plan. The generation and transmission time-series data generation process obtains the network topology, wind, solar, and load time-series data, tie-line plan, similar day tie-line power, and conventional unit generation plan for the day to be generated, solves the unit combination optimization problem, and outputs unit start-up and shutdown and active power data under specified scenario conditions, such as... Figure 6 As shown.

[0073] (1) Scheme for dismantling the connecting line For provincial power grids, the provincial dispatch center is responsible for decomposing the external power tie line plans issued by the sub-centers into node injection power plans, which are then used as boundary conditions for market operation. Therefore, when creating scenario-based samples for provincial power grids, the issue of node plan decomposition must also be addressed. Tie line plans typically provide total exchange power (i.e., "node plans") in units of "regional pairs," but actual power flow data generation and safety verification require precise power allocation down to each physical tie line. Therefore, the total power must be allocated to each line according to reasonable rules. There are two approaches to this decomposition: the historical ratio method based on similar days and the manually specified ratio method, which users can flexibly choose according to their scenario.

[0074] 1) Historical Proportioning Method Based on Similar Dates The applicable scenario is when there are no special operating requirements and it is desirable to reflect historical operating habits. For each similar day, the proportion of the total active power transmitted by each tie line member to the tie line group within the same time period is calculated. If there are multiple similar days, they are weighted according to their time proximity. Finally, the total planned tie line power for the day to be generated is allocated proportionally.

[0075] The percentage of the measured power of the i-th tie-line member relative to the total planned power on the d-th similar day is calculated using the following formula:

[0076] The weighted recommendation percentage is calculated using the following expression:

[0077] Total planned power of the generated day The proportional allocation is calculated as follows:

[0078] 2) Manually specified ratio method This system is applicable to scenarios with clear scheduling requirements, such as prioritizing the restoration of a line after maintenance or designating a channel for cross-regional transactions. By providing a web configuration page, users can set fixed allocation coefficients for each tie line group and tie line member.

[0079] (2) Unit combination The unit combination can directly call the power generation plan function to generate information based on the established unit start-up and shutdown plan, load time series data, wind and solar new energy time series data, distributed photovoltaic data, inter-provincial and regional interconnection time series data, day-ahead power trading, dispatchable load dispatch rules, etc. Considering system balance, unit operation, grid security and other constraints, it supports optimization goals such as fair dispatch, lowest system power purchase cost, energy-saving power generation dispatch, and maximization of clean energy consumption. It can complete the combination optimization of individual conventional units, gas and pumped storage units with fast start-up and shutdown, as well as the output plan of units and dispatchable loads.

[0080] The generation planning function can be directly invoked to generate unit combinations and active power time-series data. However, when generating random scenarios, due to the source-load scenario combination explosion problem, the computational workload for generating unit combinations and time-series data for each source-load scenario is enormous. The following are some implementation paths: Method 1: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios. Then, for each set of source-load random scenarios, independently call the complete power generation plan optimization model to generate unit combination and output data. Calculate the average of multiple sets of unit combination and output data to obtain the final unit power generation time series data. Method 2: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios. Then, for each set of source-load random scenarios, obtain its equivalent load time-series active power data. Cluster analysis algorithms are used to group the multiple sets of data. The median equivalent load time-series data of all data within a group is taken, and a power generation planning model is called to generate unit combinations. Then, all the original renewable energy generation time-series data and load active power data in this group are superimposed, and unit output is adjusted in conjunction with AC power flow to finally obtain the unit time-series output data for each source-load scenario. The processing method is consistent for each group. Method 3: Based on specified time and location, operational indicators, power grid planning, primary energy sources, disaster events, and meteorological conditions, generate a set of source-load operation scenarios with the highest probability and likelihood. Call the power generation planning model to obtain power generation planning data, set optimization objectives, and call the Safety Constraint Unit Combination (SCUC) to generate unit combinations. On this basis, overlay multiple sets of unconditionally random combination data to adjust unit output. This can be achieved by using safety correction with the goal of minimizing the adjustment amount, or by using the power flow imbalance power allocation function to adjust unit output. Method 4: Employing artificial intelligence, multiple sets of unconditional source-load random scenarios are generated by specifying only the time and region. A power generation planning model is then invoked, and a corresponding power generation plan is obtained for each set of source-load data. Massive amounts of data are input into a large model to solve for unit combination and active power output.

[0081] In one possible implementation, step S6 generates power flow result data based on the power distribution of injected energy sources, loads, tie lines, generator units, etc., as well as reactive power and voltage constraints, using the AC optimal power flow and conventional AC power flow functions.

[0082] By acquiring the network topology, wind-solar-load time-series data, unit combination and active power, similar day grid model and operating data for the date to be generated, and calling the power flow data generation function, a high-precision, continuous, and physically feasible power flow dataset is further generated, including the voltage amplitude, phase angle, branch power, reactive power output, etc. of each node. Figure 7 As shown. In the power flow data generation stage, conventional AC power flow solution algorithms (such as the Newton-Raphson method and the PQ fast decomposition method) can be used to achieve efficient simulation of large-scale scenarios. If there is a need for high-precision risk assessment, AC Optimal Power Flow (ACOPF) or probabilistic power flow methods can be introduced to improve the optima of the solution and the ability to characterize uncertainties.

[0083] In one possible implementation, after step S7, the rationality of the cross-sectional and time-series data at each moment is verified, and the scene data is classified according to the business requirements by selecting the appropriate indicators and eliminating redundant scenes.

[0084] Please see Figure 1 Another embodiment of the present invention also proposes a power grid operation scenario generation system, comprising: The condition setting module is used to set the scene generation conditions; The network topology generation module is used to analyze the scene generation conditions and generate the network topology. The contingency set construction module is used to construct contingency sets that adapt to different scenario generation requirements based on scenario generation conditions, and update the network topology; The wind-solar-load time-series data generation module is used to obtain the operating status of new energy power plants and loads, and combine the scene generation conditions and the network topology that adapts to the current scene generation requirements to generate wind-solar-load time-series data that conforms to the wind and solar power generation patterns and load electricity consumption patterns of the corresponding region and time period. The power generation and transmission time series data generation module is used to optimize the unit combination based on wind, solar and load time series data, and obtain the unit combination and active power under the scenario generation conditions. The power flow data generation module is used to generate power flow data based on the unit combination and active power under the scenario generation conditions. The scenario data generation module is used to associate and combine power flow data and environmental data to generate power grid operation scenario data.

[0085] like Figure 1 As shown, staff or scenario-generating intelligent agents set scenario generation conditions (including time and location, operational indicators, power grid planning and operation, primary energy supply, disaster events, meteorological conditions, dispatch plan boundaries, etc.) according to business needs. The system calls scenario data services to obtain initial basic data (including meteorological information, disaster information, primary energy supply, power grid planning, power grid model and operational data, geographic information, etc.) based on the set conditions. Then, the power grid operation scenario generation unit outputs the power grid operation scenario under the specified conditions. Finally, the generated scenario is analyzed and evaluated using functions such as simulation, indicator calculation, scenario classification and filtering to assess whether it meets business requirements. The generated scenario data is included in the big data platform asset management, and indicator tags are added to the generated scenario data and the initial data.

[0086] In one possible implementation, the scenario generation conditions set by the condition setting module include any one or more combinations of time and location, operating indicators, power grid planning, primary energy, disaster events, meteorological environment, dispatching plans, and load regulation capabilities; wherein, time and location are mandatory, and other conditions are optional. The time and location settings specify the time period, data sampling interval, and power grid range. Based on the specified time, the system extracts the matching power grid model version and power grid operation data. For historical moments, it retrieves the corresponding historical power grid data, and for future moments, it retrieves the corresponding planned forecast data. If historical data or planned forecast data is missing, it combines other settings to retrieve similar daily power grid operation data. When only the time and location conditions are specified, the system reads the specified power grid model and the operation data for the specified time period through the scenario data service. The operational indicators are set as target values ​​for the generated power grid operation scenarios. These indicators include categories such as safety and stability, balance and regulation, renewable energy consumption, and power grid overview. Among them, the safety and stability category sets static safety indicators, the balance and regulation category includes balance capacity and regulation capacity, the renewable energy consumption category is renewable energy obstruction indicators, and the power grid overview indicators include AC / DC channel disaster values ​​and levels, as well as the scale of equipment failures under disasters. Based on the operational indicators, similar daily power grid models and operational data are obtained through indicator label matching. With the goal of minimizing the adjustment amount, the indicators are converted into power grid generation and consumption power and network channel power. The power grid planning settings include new and expanded power plants, substations, new energy power stations, distributed photovoltaics, AC / DC lines, transformers, various reactive power compensations, and changes on all sides of the power grid, including generation, transmission, transformation, distribution, and consumption. Based on the settings, the system checks whether the power grid planning equipment has been modeled in the existing model version. If it has been modeled, it can be used directly; if it has not been modeled, a new model should be created in the future state model. Primary energy settings include coal storage, gas storage, and water inflow conditions, which convert relevant data into the total electricity that the power plant can generate. The disaster event settings include the disaster event type and the time and location of the event; based on the disaster event, using the anticipated faults that take into account external disasters, the disaster event is transformed into a change in the power grid operation status. On the initial power grid operation data extracted or generated by the scenario, faults and power fluctuations are superimposed to generate power grid operation mode data under the disaster scenario; The meteorological environment settings include meteorological type settings. The meteorological environment time series data and time and location settings are matched to conform to the climate patterns of the external environment of the power grid at the specified time and region. Based on the set meteorological environment time series data, the matching daily meteorological data and operation mode data of the power grid are obtained, and combined with the source and load time series data to generate and transform the set conditions into active power time series data of new energy, distributed photovoltaic and load. The dispatch plan sets price factors, reserve constraints, medium- and long-term transactions, spot transactions, maintenance plans, and tie-line plan conditions. It determines whether the specified time period is a historical or future time. If it is a historical time, it generates active power generation data and tie-line exchange power data based on historical planned forecasts and actual operating data, according to the settings and in conjunction with the dispatch plan function. If it is a future time and planned forecast data is available, it adjusts the data based on the planned forecast data, according to the set conditions, and in conjunction with the dispatch plan function, and extrapolates future trends. If it is a future time but planned forecast data is missing, it generates planned forecast data using the dispatch plan function and extrapolates future trends. The load regulation capacity is set according to the power supply area of ​​equivalent load, including the load scale and its adjustable or interruptible capacity of industrial users, the load scale and its adjustable capacity of commercial and public building temperature control and lighting, the load scale and its adjustable capacity of residential users, the load scale and its adjustable capacity of electric vehicles, and the scale and its adjustable capacity of aggregators. Based on the conditions, the adjustable capacity is quantified to generate load response speed, regulation power and capacity, response direction and control method.

[0087] In one possible implementation, the power grid operation scenario generation system of this invention further includes a condition verification module, which is used to perform conflict verification on the set scenario generation conditions. The conflict verification includes the degree of matching between time, location and meteorological environment, disaster events, dispatch plans, power grid planning, primary energy, operation indicators and load regulation capabilities. Based on the analysis and statistics of the qualitative or quantitative correlation between various scenario generation conditions, verification rules are formed.

[0088] In one possible implementation, the network topology generation module obtains the power grid model and the on / off status or switching status of the equipment, and performs topology analysis to generate the power grid network topology and corresponding calculation model under the scenario generation conditions.

[0089] In one possible implementation, the anticipatory fault set construction module acquires preset external meteorological conditions, equipment parameters, meteorological data during historical similar fault periods, and power grid operation data, establishes a probability model for single equipment faults, generates a clustered combined fault set, and outputs the IDs of all faulted equipment in the fault set under the corresponding meteorological conditions.

[0090] In one possible implementation, when generating a cluster of combined fault sets, the anticipated fault set construction module dynamically assesses the probability of power grid equipment failure caused by the external environment over a period of time, taking into account the security and stability assessment needs of different business applications, external environment, geographical location, and equipment design parameters, and generates anticipated fault sets adapted to different scenario generation needs; it analyzes the propagation and evolution process of cascading failures caused by a single fault, filters out cascading failure types and occurrence scenarios, and generates a cluster of anticipated fault sets for power grid failures over a period of time; considering the generation of anticipated fault sets based on natural environmental factors, it starts from the mechanism of natural disasters affecting equipment, establishes a model of the probability of equipment failure caused by natural disasters, and then calculates the probability of clustered anticipated faults based on the assessment of the probability of single equipment failure caused by disasters, and generates the anticipated fault set.

[0091] In one possible implementation, the wind-solar-load time-series data generation module generates a set of deterministic expected scenarios or multiple sets of random scenario source-load time-series data within the feasible domain, based on scenario generation conditions. The set of time-series data includes centralized wind power and photovoltaic (PV) and distributed PV output data, load forecast data, and generates several data entries for wind farms, PV stations, distributed PV, and bus loads based on the number of new energy power plants, equivalent loads, and equivalent generating units. It also includes identical centralized wind-solar output data, distributed PV data, and load data, and can decompose multiple sets of new energy power plants, equivalent loads, and equivalent generating unit data. Multiple sets of time-series data include multiple sets of feasible domain time-series data that conform to the fluctuation patterns of wind power, PV, distributed PV, and load within specified conditions. The generated wind-solar-load time-series data conforms to the inherent correlation patterns of various factors and wind-solar output and load electricity consumption within the specified region and time period, and conforms to the fluctuation patterns of wind-solar output and load within the specified region and time period.

[0092] In one possible implementation, the wind-solar-load time-series data generation module generates wind-solar-load time-series data with a minimum time length of 1 hour and a minimum time interval of 1 minute. It generates time-series data for 1 to 24 hours at a time, with time intervals of 1, 5, 15, or 60 minutes, supporting the generation needs of time-series data with different durations and time intervals. For data with a weekly duration, the time interval is no more than 15 minutes, and for data with a monthly or yearly duration, the time interval is 60 minutes. When it is necessary to generate time-series data for the next day, the data is generated on a daily basis, using the results generated on the current day as input for the data generated on the next day, and so on in a rolling manner. When the wind and solar power generation or load power consumption within a specified period is specified, the generated time-series data meets the power consumption constraints.

[0093] In one possible implementation, the power generation and transmission time series data generation module acquires the network topology, wind and solar load time series data, grid plan, tie line power of similar days, and conventional unit power generation plan for the day to be generated, solves the unit combination optimization problem, and outputs the unit start-up and shutdown mode and active power data under the specified scenario generation conditions. When solving the unit combination optimization problem, the provincial power grid is decomposed into sub-plans, and the total power is distributed to each line using the historical ratio method based on similar days or the manually specified ratio method. The total power is broken down into individual lines based on the historical proportional method for similar days, including the following steps: For each similar day, the total active power transmitted by each tie line member within the same time period is counted as a percentage of the tie line group. If multiple similar days exist, the total planned tie line power for the day to be generated is allocated proportionally based on time proximity. The percentage of the measured power of the i-th tie line member on the d-th similar day to the total planned power of the tie line is calculated using the following formula:

[0094] The weighted recommendation percentage is calculated using the following expression:

[0095] Total planned power of the generated day The proportional allocation is calculated as follows:

[0096] The manual allocation ratio method involves manually setting a fixed allocation coefficient for each tie line group and tie line member. The implementation methods include the following: Method 1: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios; for each set of source-load random scenarios, independently call the complete power generation plan optimization model to generate unit combination and output data, and calculate the average of multiple sets of unit combination and output data to obtain unit power generation time series data. Method 2: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios. For each set of source-load random scenarios, obtain the corresponding equivalent load time-series active power data. Use a clustering analysis algorithm to cluster the multiple sets of data into groups. Take the median equivalent load time-series data of all data in the group and call the power generation planning model to generate unit combinations. Then, superimpose all the original new energy power generation time-series data and load active power data in the corresponding group, and adjust the unit output in combination with AC power flow to obtain the unit time-series output data for each source-load scenario. The processing method is the same for each group. Method 3: Based on the specified time and location, operating indicators, power grid planning, primary energy sources, disaster events, or meteorological conditions, generate the most probable set of source-load operating scenarios; call the power generation planning model to obtain power generation planning data, set optimization objectives, and call the Safety Constraint Unit Combination (SCUC) to generate unit combinations; on this basis, superimpose multiple sets of unconditional random combination data to adjust unit output; Method 4: Using artificial intelligence, multiple sets of unconditional source-load random scenarios are generated by specifying only the time and region, and the power generation planning model is called. For each set of source-load data, a corresponding power generation plan is obtained, and the unit combination and active power output are solved through a large model.

[0097] In one possible implementation, the power flow data generation module generates a power flow dataset by acquiring network topology, wind-solar-load time-series data, unit combination and active power, similar day grid model and operation data under the conditions of the scenario to be generated; the power flow data includes the voltage amplitude, phase angle, branch power and reactive power of each node.

[0098] In one possible implementation, the power grid operation scenario generation system of this embodiment further includes a scenario verification module, which is used to verify the rationality of the cross-section and time series data at each moment after generating power grid operation scenario data, select corresponding indicators to classify the scenario data according to business needs, and eliminate redundant scenarios.

[0099] Another embodiment of the present invention also proposes an electronic device, including a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the power grid operation scenario generation method described above.

[0100] Another embodiment of the present invention provides a computer-readable storage medium storing at least one instruction that, when executed by a processor, implements the power grid operation scenario generation method.

[0101] The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals. For ease of explanation, the above content only shows the parts related to the embodiments of the present invention; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. This computer-readable storage medium is non-transitory and can be stored in storage devices formed by various electronic devices, enabling the execution process described in the method of the embodiments of the present invention.

[0102] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0104] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for generating power grid operation scenarios, characterized in that, include: Set the scene generation conditions; Analyze the scenario generation conditions and generate the network topology; Based on the scenario generation conditions, construct a set of anticipated faults that adapt to the generation requirements of different scenarios, and update the network topology; The system acquires the operating status of new energy power plants and loads, and combines the scenario generation conditions with the network topology that adapts to the current scenario generation requirements to generate wind and solar load time-series data that conforms to the wind and solar power generation patterns and load electricity consumption patterns of the corresponding region and time period. Based on wind, solar and load time series data, unit combination optimization is performed to obtain unit combination and active power under the generated scenario conditions; Power flow data is generated based on the unit combination and active power under the scenario generation conditions. By linking and combining power flow data and environmental data, power grid operation scenario data can be generated.

2. The method for generating power grid operation scenarios according to claim 1, characterized in that, In the step of setting scenario generation conditions, the scenario generation conditions include any one or more combinations of time and location, operating indicators, power grid planning, primary energy, disaster events, meteorological environment, dispatching plan, and load regulation capacity; Among them, time and location are mandatory, while other conditions are optional; The time and location settings specify the time period, data sampling interval, and power grid range. Based on the specified time, the system extracts the matching power grid model version and power grid operation data. For historical moments, it retrieves the corresponding historical power grid data, and for future moments, it retrieves the corresponding planned forecast data. If historical data or planned forecast data is missing, it combines other settings to retrieve similar daily power grid operation data. When only the time and location conditions are specified, the system reads the specified power grid model and the operation data for the specified time period through the scenario data service. The operational indicators are set as target values ​​for the generated power grid operation scenarios. These indicators include categories such as safety and stability, balance and regulation, renewable energy consumption, and power grid overview. Among them, the safety and stability category sets static safety indicators, the balance and regulation category includes balance capacity and regulation capacity, the renewable energy consumption category is renewable energy obstruction indicators, and the power grid overview indicators include AC / DC channel disaster values ​​and levels, as well as the scale of equipment failures under disasters. Based on the operational indicators, similar daily power grid models and operational data are obtained through indicator label matching. With the goal of minimizing the adjustment amount, the indicators are converted into power grid generation and consumption power and network channel power. The power grid planning settings include new and expanded power plants, substations, new energy power stations, distributed photovoltaics, AC / DC lines, transformers, various reactive power compensations, and changes on all sides of the power grid, including generation, transmission, transformation, distribution, and consumption. Based on the settings, the system checks whether the power grid planning equipment has been modeled in the existing model version. If it has been modeled, it can be used directly; if it has not been modeled, a new model should be created in the future state model. Primary energy settings include coal storage, gas storage, and water inflow conditions, which convert relevant data into the total electricity that the power plant can generate. The disaster event settings include the disaster event type and the time and location of the event; based on the disaster event, using the anticipated faults that take into account external disasters, the disaster event is transformed into a change in the power grid operation status. On the initial power grid operation data extracted or generated by the scenario, faults and power fluctuations are superimposed to generate power grid operation mode data under the disaster scenario; The meteorological environment settings include meteorological type settings. The meteorological environment time series data and time and location settings are matched to conform to the climate patterns of the external environment of the power grid at the specified time and region. Based on the set meteorological environment time series data, the matching daily meteorological data and operation mode data of the power grid are obtained, and combined with the source and load time series data to generate and transform the set conditions into active power time series data of new energy, distributed photovoltaic and load. The dispatch plan sets price factors, reserve constraints, medium- and long-term transactions, spot transactions, maintenance plans, and tie-line plan conditions. It determines whether the specified time period is a historical or future time. If it is a historical time, it generates active power generation data and tie-line exchange power data based on historical planned forecasts and actual operating data, according to the settings and in conjunction with the dispatch plan function. If it is a future time and planned forecast data is available, it adjusts the data based on the planned forecast data, according to the set conditions, and in conjunction with the dispatch plan function, and extrapolates future trends. If it is a future time but planned forecast data is missing, it generates planned forecast data using the dispatch plan function and extrapolates future trends. The load regulation capacity is set according to the power supply area of ​​equivalent load, including the load scale and its adjustable or interruptible capacity of industrial users, the load scale and its adjustable capacity of commercial and public building temperature control and lighting, the load scale and its adjustable capacity of residential users, the load scale and its adjustable capacity of electric vehicles, and the scale and its adjustable capacity of aggregators. Based on the conditions, the adjustable capacity is quantified to generate load response speed, regulation power and capacity, response direction and control method.

3. The method for generating power grid operation scenarios according to claim 2, characterized in that, It also includes a step of conflict verification of the set scenario generation conditions. Conflict verification includes the degree of matching between time, location and meteorological environment, disaster events, dispatch plans, power grid planning, primary energy, operating indicators and load regulation capabilities. Based on the analysis and statistics of the qualitative or quantitative correlation between various scenario generation conditions, verification rules are formed.

4. The method for generating power grid operation scenarios according to claim 1, characterized in that, In the step of generating network topology under the analysis scenario generation conditions, the power grid model and the on / off status or switching status of the equipment are obtained, and topology analysis is performed to generate the power grid network topology and corresponding calculation model under the scenario generation conditions.

5. The method for generating power grid operation scenarios according to claim 1, characterized in that, In the step of constructing a set of anticipated faults to meet the generation requirements of different scenarios based on the scenario generation conditions and updating the network topology, preset external meteorological conditions, equipment parameters, meteorological data during similar historical faults and power grid operation data are obtained. A probability model is established for single equipment faults, a set of clustered combined faults is generated, and the IDs of all faulted equipment in the fault set under the corresponding meteorological conditions are output.

6. The method for generating power grid operation scenarios according to claim 5, characterized in that, In the step of generating a set of clustered combined faults, based on the security and stability assessment needs of different business applications, and taking into account the external environment, geographical location, and equipment design parameters, the probability of power grid equipment failure caused by the external environment within a certain period of time is dynamically assessed, and a set of anticipated faults that adapts to the generation needs of different scenarios is generated. The propagation and evolution of cascading failures caused by a single fault are analyzed, and the types and scenarios of cascading failures are screened out to generate a set of anticipated power grid failures over a period of time. The generation of the anticipated failure set considering natural environmental factors starts from the mechanism of natural disasters affecting equipment, establishes a model of the probability of equipment failure caused by natural disasters, and then calculates the probability of clustered anticipated failures based on the probability assessment of single equipment failures caused by disasters to generate the anticipated failure set.

7. The method for generating power grid operation scenarios according to claim 1, characterized in that, When acquiring the operating status of new energy power plants and loads, and combining the scenario generation conditions and network topology adapted to the current scenario generation requirements, to generate wind, solar, and load time-series data that conforms to the wind and solar power generation patterns and load consumption patterns of the corresponding region and time period, according to the scenario generation conditions, a set of deterministic expected scenarios or multiple sets of random scenario source-load time-series data within the feasible domain are generated. One set of time-series data includes centralized wind power and photovoltaic and distributed photovoltaic power output data and load forecast data within the corresponding region. Several sets of wind farm, photovoltaic station, distributed photovoltaic, and bus load data are generated according to the number of new energy power plants, equivalent loads, and equivalent generating units, as well as the same centralized wind and solar power output data, distributed photovoltaic data, and load data, and multiple sets of new energy power plants, equivalent loads, and equivalent generating unit data can be decomposed. Multiple sets of time-series data include multiple sets of time-series data within the feasible domain that conform to the fluctuation patterns of wind power, photovoltaic, distributed photovoltaic, and load within specified conditions. The generated wind, solar, and load time-series data conforms to the inherent correlation patterns of various factors and wind and solar power output and load consumption within the specified region and time period, and conforms to the fluctuation patterns of wind and solar power output and load within the specified region and time period.

8. The method for generating power grid operation scenarios according to claim 7, characterized in that, The minimum time unit for the wind, solar, and load time series data is 1 hour, and the minimum time interval is 1 minute. Time series data for 1 to 24 hours can be generated at a time, with time intervals of 1, 5, 15, or 60 minutes, supporting the generation of time series data with different durations and time intervals. For data with a weekly duration, the time interval is no more than 15 minutes, and for data with a monthly or yearly duration, the time interval is 60 minutes. When it is necessary to generate time series data for the next day, the data is generated on a daily basis, using the results generated on the current day as input for the data generated on the next day, and so on in a rolling manner. When the wind and solar power generation or load power consumption within a specified period is specified, the generated time series data meets the power consumption constraints.

9. The method for generating power grid operation scenarios according to claim 1, characterized in that, The unit combination optimization based on wind, solar and load time series data is performed. When obtaining the unit combination and active power under the scenario generation conditions, the network topology, wind, solar and load time series data, port plan, tie line power of similar days, and conventional unit power generation plan of the day to be generated are obtained. The unit combination optimization problem is solved, and the unit start-up and shutdown mode and active power data under the specified scenario generation conditions are output. In the steps of solving the unit combination optimization problem, the provincial power grid is decomposed into sub-plans, and the total power is distributed to each line using the historical proportion method based on similar days or the manually specified proportion method. The method of allocating total power to each line based on the historical proportion of similar days includes the following steps: For each similar day, the total active power transmitted by each tie line member within the same time period is counted as a percentage of the tie line group. If multiple similar days exist, the total planned tie line power for the day to be generated is allocated proportionally based on time proximity. The percentage of the measured power of the i-th tie line member on the d-th similar day to the total planned power of the tie line is calculated using the following formula: The weighted recommendation percentage is calculated using the following expression: Total planned power of the day to be generated The proportional allocation is calculated as follows: The manual allocation ratio method involves manually setting a fixed allocation coefficient for each tie line group and tie line member. The implementation methods include the following: Method 1: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios; for each set of source-load random scenarios, independently call the complete power generation plan optimization model to generate unit combination and output data, and calculate the average of multiple sets of unit combination and output data to obtain unit power generation time series data. Method 2: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios. For each set of source-load random scenarios, obtain the corresponding equivalent load time-series active power data. Use a clustering analysis algorithm to cluster the multiple sets of data into groups. Take the median equivalent load time-series data of all data in the group and call the power generation planning model to generate unit combinations. Then, superimpose all the original new energy power generation time-series data and load active power data in the corresponding group, and adjust the unit output in combination with AC power flow to obtain the unit time-series output data for each source-load scenario. The processing method is the same for each group. Method 3: Based on the specified time and location, operating indicators, power grid planning, primary energy sources, disaster events, or meteorological conditions, generate the most probable set of source-load operating scenarios; call the power generation planning model to obtain power generation planning data, set optimization objectives, and call the Safety Constraint Unit Combination (SCUC) to generate unit combinations; on this basis, superimpose multiple sets of unconditional random combination data to adjust unit output; Method 4: Using artificial intelligence, multiple sets of unconditional source-load random scenarios are generated by specifying only the time and region, and the power generation planning model is called. For each set of source-load data, a corresponding power generation plan is obtained, and the unit combination and active power output are solved through a large model.

10. The method for generating power grid operation scenarios according to claim 1, characterized in that, In the step of generating power flow data based on the unit combination and active power under the scenario generation conditions, a power flow dataset is generated by acquiring the network topology, wind-solar-load time-series data, unit combination and active power, similar day grid model and operation data under the scenario generation conditions of the day to be generated; the power flow data includes the voltage amplitude, phase angle, branch power and reactive power of each node.

11. The method for generating power grid operation scenarios according to claim 1, characterized in that, It also includes verifying the rationality of the cross-section and time series data at each moment after generating power grid operation scenario data, selecting appropriate indicators to classify the scenario data according to business needs, and eliminating redundant scenarios.

12. A power grid operation scenario generation system, characterized in that, include: The condition setting module is used to set the scene generation conditions; The network topology generation module is used to analyze the scene generation conditions and generate the network topology. The contingency set construction module is used to construct contingency sets that adapt to different scenario generation requirements based on scenario generation conditions, and update the network topology; The wind-solar-load time-series data generation module is used to obtain the operating status of new energy power plants and loads, and combine the scene generation conditions and the network topology that adapts to the current scene generation requirements to generate wind-solar-load time-series data that conforms to the wind and solar power generation patterns and load electricity consumption patterns of the corresponding region and time period. The power generation and transmission time series data generation module is used to optimize the unit combination based on wind, solar and load time series data, and obtain the unit combination and active power under the scenario generation conditions. The power flow data generation module is used to generate power flow data based on the unit combination and active power under the scenario generation conditions. The scenario data generation module is used to associate and combine power flow data and environmental data to generate power grid operation scenario data.

13. The power grid operation scenario generation system according to claim 12, characterized in that, The scenario generation conditions set by the condition setting module include any one or more combinations of time and location, operating indicators, power grid planning, primary energy, disaster events, meteorological environment, dispatching plan, and load regulation capacity; Among them, time and location are mandatory, while other conditions are optional; The time and location settings specify the time period, data sampling interval, and power grid range. Based on the specified time, the system extracts the matching power grid model version and power grid operation data. For historical moments, it retrieves the corresponding historical power grid data, and for future moments, it retrieves the corresponding planned forecast data. If historical data or planned forecast data is missing, it combines other settings to retrieve similar daily power grid operation data. When only the time and location conditions are specified, the system reads the specified power grid model and the operation data for the specified time period through the scenario data service. The operational indicators are set as target values ​​for the generated power grid operation scenarios. These indicators include categories such as safety and stability, balance and regulation, renewable energy consumption, and power grid overview. Among them, the safety and stability category sets static safety indicators, the balance and regulation category includes balance capacity and regulation capacity, the renewable energy consumption category is renewable energy obstruction indicators, and the power grid overview indicators include AC / DC channel disaster values ​​and levels, as well as the scale of equipment failures under disasters. Based on the operational indicators, similar daily power grid models and operational data are obtained through indicator label matching. With the goal of minimizing the adjustment amount, the indicators are converted into power grid generation and consumption power and network channel power. The power grid planning settings include new and expanded power plants, substations, new energy power stations, distributed photovoltaics, AC / DC lines, transformers, various reactive power compensations, and changes on all sides of the power grid, including generation, transmission, transformation, distribution, and consumption. Based on the settings, the system checks whether the power grid planning equipment has been modeled in the existing model version. If it has been modeled, it can be used directly; if it has not been modeled, a new model should be created in the future state model. Primary energy settings include coal storage, gas storage, and water inflow conditions, which convert relevant data into the total electricity that the power plant can generate. The disaster event settings include the disaster event type and the time and location of the event; based on the disaster event, using the anticipated faults that take into account external disasters, the disaster event is transformed into a change in the power grid operation status. On the initial power grid operation data extracted or generated by the scenario, faults and power fluctuations are superimposed to generate power grid operation mode data under the disaster scenario; The meteorological environment settings include meteorological type settings. The meteorological environment time series data and time and location settings are matched to conform to the climate patterns of the external environment of the power grid at the specified time and region. Based on the set meteorological environment time series data, the matching daily meteorological data and operation mode data of the power grid are obtained, and combined with the source and load time series data to generate and transform the set conditions into active power time series data of new energy, distributed photovoltaic and load. The dispatch plan sets price factors, reserve constraints, medium- and long-term transactions, spot transactions, maintenance plans, and tie-line plan conditions. It determines whether the specified time period is a historical or future time. If it is a historical time, it generates active power generation data and tie-line exchange power data based on historical planned forecasts and actual operating data, according to the settings and in conjunction with the dispatch plan function. If it is a future time and planned forecast data is available, it adjusts the data based on the planned forecast data, according to the set conditions, and in conjunction with the dispatch plan function, and extrapolates future trends. If it is a future time but planned forecast data is missing, it generates planned forecast data using the dispatch plan function and extrapolates future trends. The load regulation capacity is set according to the power supply area of ​​equivalent load, including the load scale and its adjustable or interruptible capacity of industrial users, the load scale and its adjustable capacity of commercial and public building temperature control and lighting, the load scale and its adjustable capacity of residential users, the load scale and its adjustable capacity of electric vehicles, and the scale and its adjustable capacity of aggregators. Based on the conditions, the adjustable capacity is quantified to generate load response speed, regulation power and capacity, response direction and control method.

14. The power grid operation scenario generation system according to claim 13, characterized in that, It also includes a condition verification module, which is used to perform conflict verification on the set scenario generation conditions. Conflict verification includes the degree of matching between time, location and meteorological environment, disaster events, scheduling plans, power grid planning, primary energy, operating indicators and load regulation capabilities. Based on the analysis and statistics of the qualitative or quantitative correlation between various scenario generation conditions, verification rules are formed.

15. The power grid operation scenario generation system according to claim 12, characterized in that, The network topology generation module obtains the power grid model and the on / off status or switching status of the equipment, and performs topology analysis to generate the power grid network topology and corresponding calculation model under the scenario generation conditions.

16. The power grid operation scenario generation system according to claim 12, characterized in that, The anticipated fault set construction module acquires preset external meteorological conditions, equipment parameters, meteorological data during historical similar fault periods, and power grid operation data. It establishes a probability model for single equipment faults, generates a clustered combined fault set, and outputs the IDs of all faulty equipment in the fault set under the corresponding meteorological conditions.

17. The power grid operation scenario generation system according to claim 16, characterized in that, When generating a cluster of combined fault sets, the contingent fault set construction module dynamically assesses the probability of power grid equipment failure caused by the external environment within a certain period of time, taking into account the security and stability assessment needs of different business applications, combined with the external environment, geographical location and equipment design parameters, and generates contingent fault sets that adapt to the generation needs of different scenarios. The propagation and evolution of cascading failures caused by a single fault are analyzed, and the types and scenarios of cascading failures are screened out to generate a set of anticipated power grid failures over a period of time. The generation of the anticipated failure set considering natural environmental factors starts from the mechanism of natural disasters affecting equipment, establishes a model of the probability of equipment failure caused by natural disasters, and then calculates the probability of clustered anticipated failures based on the probability assessment of single equipment failures caused by disasters to generate the anticipated failure set.

18. The power grid operation scenario generation system according to claim 12, characterized in that, The wind-solar-load time-series data generation module generates a set of deterministic expected scenarios or multiple sets of random scenario source-load time-series data within the feasible domain, based on scenario generation conditions. A set of time-series data includes centralized wind power and photovoltaic (PV) and distributed PV output data, load forecast data, and generates several data entries for wind farms, PV stations, distributed PV, and bus loads based on the number of new energy power plants, equivalent loads, and equivalent generating units. It also generates identical centralized wind-solar output data, distributed PV data, and load data, and can decompose multiple sets of new energy power plants, equivalent loads, and equivalent generating unit data. Multiple sets of time-series data include multiple sets of feasible domain time-series data that conform to the fluctuation patterns of wind power, PV, distributed PV, and load within specified conditions. The generated wind-solar-load time-series data conforms to the inherent correlation patterns of various factors and wind-solar output and load electricity consumption within the specified region and time period, and conforms to the fluctuation patterns of wind-solar output and load within the specified region and time period.

19. The power grid operation scenario generation system according to claim 18, characterized in that, The wind-solar-load time-series data generation module generates time-series data with a minimum time length of 1 hour and a minimum time interval of 1 minute. It generates time-series data for 1 to 24 hours at a time, with time intervals of 1, 5, 15, or 60 minutes, supporting the generation of time-series data of different lengths and time intervals. For data with a weekly time length, the time interval is no more than 15 minutes; for data with a monthly or yearly time length, the time interval is 60 minutes. When it is necessary to generate time-series data for the next day, the data is generated on a daily basis, using the results generated on the current day as input for the data generated on the next day, and so on in a rolling manner. When the wind and solar power generation or load power consumption within a specified period is specified, the generated time-series data meets the power consumption constraints.

20. The power grid operation scenario generation system according to claim 12, characterized in that, The power generation and transmission time series data generation module acquires the network topology, wind and solar load time series data, grid plan, tie line power of similar days, and conventional unit power generation plan for the day to be generated, solves the unit combination optimization problem, and outputs the unit start-up and shutdown mode and active power data under the specified scenario generation conditions. When solving the unit combination optimization problem, the provincial power grid is decomposed into sub-plans, and the total power is distributed to each line using the historical ratio method based on similar days or the manually specified ratio method. The method of allocating total power to each line based on the historical proportion of similar days includes the following steps: For each similar day, the total active power transmitted by each tie line member within the same time period is counted as a percentage of the tie line group. If multiple similar days exist, the total planned tie line power for the day to be generated is allocated proportionally based on time proximity. The percentage of the measured power of the i-th tie line member on the d-th similar day to the total planned power of the tie line is calculated using the following formula: The weighted recommendation percentage is calculated using the following expression: Total planned power of the day to be generated The proportional allocation is calculated as follows: The manual allocation ratio method involves manually setting a fixed allocation coefficient for each tie line group and tie line member. The implementation methods include the following: Method 1: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios; for each set of source-load random scenarios, independently call the complete power generation plan optimization model to generate unit combination and output data, and calculate the average of multiple sets of unit combination and output data to obtain unit power generation time series data. Method 2: Specify only the time and region to generate multiple sets of unconditional source-load random scenarios. For each set of source-load random scenarios, obtain the corresponding equivalent load time-series active power data. Use a clustering analysis algorithm to cluster the multiple sets of data into groups. Take the median equivalent load time-series data of all data in the group and call the power generation planning model to generate unit combinations. Then, superimpose all the original new energy power generation time-series data and load active power data in the corresponding group, and adjust the unit output in combination with AC power flow to obtain the unit time-series output data for each source-load scenario. The processing method is the same for each group. Method 3: Based on the specified time and location, operating indicators, power grid planning, primary energy sources, disaster events, or meteorological conditions, generate the most probable set of source-load operating scenarios; call the power generation planning model to obtain power generation planning data, set optimization objectives, and call the Safety Constraint Unit Combination (SCUC) to generate unit combinations; on this basis, superimpose multiple sets of unconditional random combination data to adjust unit output; Method 4: Using artificial intelligence, multiple sets of unconditional source-load random scenarios are generated by specifying only the time and region, and the power generation planning model is called. For each set of source-load data, a corresponding power generation plan is obtained, and the unit combination and active power output are solved through a large model.

21. The power grid operation scenario generation system according to claim 12, characterized in that, The power flow data generation module generates a power flow dataset by acquiring network topology, wind-solar-load time-series data, unit combination and active power, similar day grid model and operation data under the scenario generation conditions of the day to be generated; the power flow data includes the voltage amplitude, phase angle, branch power and reactive power of each node.

22. The power grid operation scenario generation system according to claim 12, characterized in that, It also includes a scenario verification module, which is used to verify the rationality of the cross section and time series data at each moment after generating power grid operation scenario data, select appropriate indicators to classify the scenario data according to business needs, and remove redundant scenarios.

23. An electronic device, characterized in that, It includes a processor and a memory, the processor being used to execute a computer program stored in the memory to implement the power grid operation scenario generation method as described in any one of claims 1 to 11.

24. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, which, when executed by a processor, implements the power grid operation scenario generation method as described in any one of claims 1 to 11.

Images