Adaptive scheduling method, system and device for multi-energy power grid
By collecting and predicting time-series environmental data and using game theory intelligent agents for clean energy scheduling, the problems of volatility and uncertainty in clean energy power generation have been solved, and efficient and stable operation of multi-energy power grids has been achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 国网江苏省电力有限公司睢宁县供电分公司
- Filing Date
- 2026-03-13
- Publication Date
- 2026-06-05
AI Technical Summary
Existing grid dispatching methods cannot effectively cope with the volatility and uncertainty of clean energy power generation, resulting in low dispatching efficiency and poor power supply stability of multi-energy grids.
By collecting and predicting time-series environmental data, a predictive time-series environmental dataset is established, generating power generation prediction results with trust level labels. Power supply demand is decomposed into multiple sub-demands. A game theory agent is used to conduct time-series adaptation game with the power generation prediction results, configuring clean energy dispatch and compensating for unconfigured demand.
It enables efficient coordinated dispatch of clean energy and traditional energy in multi-energy power grids, improving the dispatch efficiency and power supply stability of the energy grid.
Smart Images
Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] This invention relates to the field of power grid dispatching technology, specifically to adaptive dispatching methods, systems, and equipment for multi-energy power grids. Background Technology
[0002] With the rapid development of renewable energy, traditional grid dispatching methods are struggling to address the challenge of efficient coordinated dispatching of clean and traditional energy sources in multi-energy grids. The volatility and uncertainty of renewable energy generation, such as wind and solar power, complicates the matching of grid supply capacity with load demand. Traditional grid dispatching relies on fixed generation forecasts and supply demand, failing to consider environmental factors and real-time data changes, resulting in low dispatching efficiency, significant energy waste, and poor system reliability. Current solutions typically rely on relatively simple models, failing to effectively integrate real-time environmental data, and still have shortcomings in optimizing dispatching to accommodate the volatility of clean energy generation and load demand fluctuations. Summary of the Invention
[0003] This application provides an adaptive dispatching method, system, and equipment for multi-energy power grids, which addresses the technical problem that existing power grid dispatching methods cannot effectively cope with the volatility and uncertainty of clean energy power generation, resulting in low dispatching efficiency and poor power supply stability in multi-energy power grids.
[0004] The first aspect of this application provides an adaptive scheduling method for a multi-energy power grid. The method includes: after performing time-series environmental data acquisition, establishing a predicted time-series environmental dataset based on the acquired data; using the predicted time-series environmental dataset to predict energy generation in the multi-energy power grid, establishing an energy generation prediction result, wherein the energy generation prediction result is assigned a trust level identifier; decomposing the power supply demand into N sub-power supply demands, and using each sub-power supply demand as a game agent to perform a time-series adaptation game with the energy generation prediction result; configuring the scheduling of clean energy sub-power supply demands based on the time-series adaptation game result, and establishing compensation scheduling for unconfigured sub-power supply demands; and performing adaptive scheduling management based on the sub-power supply demand scheduling and the compensation scheduling.
[0005] A second aspect of this application provides an adaptive scheduling system for a multi-energy power grid, the system comprising: a time-series environmental data prediction module, which, after performing time-series environmental data acquisition, establishes a predicted time-series environmental dataset based on the time-series environmental data acquisition results; an energy generation prediction module, which uses the predicted time-series environmental dataset to predict energy generation in the multi-energy power grid and establish an energy generation prediction result, the energy generation prediction result being set with a trust level identifier; a time-series adaptation game module, which decomposes the power supply demand into N sub-power supply demands, and uses each sub-power supply demand as a game agent to perform a time-series adaptation game with the energy generation prediction result; a scheduling demand configuration module, which configures the scheduling of clean energy sub-power supply demands based on the time-series adaptation game results, and establishes compensation scheduling for unconfigured sub-power supply demands; and an adaptive scheduling management module, which performs adaptive scheduling management based on the sub-power supply demand scheduling and the compensation scheduling.
[0006] A third aspect of this application provides an electronic device comprising: a processor coupled to a memory for storing a program that, when executed by the processor, causes the system to perform the method described in any of the first aspects.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0008] The adaptive scheduling method, system, and equipment for multi-energy power grids provided in this application relate to the field of power grid scheduling technology. By collecting and predicting time-series environmental data, a predicted time-series environmental dataset is established, generating power generation prediction results with trust level labels. Power supply demand is decomposed into multiple sub-demands. A game-theoretic intelligent agent is used to conduct time-series adaptation game with the power generation prediction results to configure clean energy scheduling and compensate for unconfigured demand. This solves the technical problem that existing power grid scheduling methods cannot effectively cope with the volatility and uncertainty of clean energy power generation, leading to low scheduling efficiency and poor power supply stability in multi-energy power grids. It achieves efficient collaborative scheduling of clean energy and traditional energy in multi-energy power grids through time-series environmental data prediction and game-theoretic intelligent agent scheduling, thereby improving the scheduling efficiency and power supply stability of the power grid. Attached Figure Description
[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0010] Figure 1 A schematic flowchart of the adaptive scheduling method for multi-energy power grids provided in this application embodiment;
[0011] Figure 2 This is a schematic diagram of the adaptive dispatch system structure for a multi-energy power grid provided in an embodiment of this application;
[0012] Figure 3 This application provides a schematic diagram of the structure of an electronic device.
[0013] Figure labeling: Time series environmental data prediction module 11, energy generation prediction module 12, time series adaptation game module 13, scheduling demand configuration module 14, adaptive scheduling management module 15, electronic device 300, memory 301, processor 302, communication interface 303, bus architecture 304. Detailed Implementation
[0014] This application provides an adaptive dispatching method, system, and equipment for multi-energy power grids, which addresses the technical problem that existing power grid dispatching methods cannot effectively cope with the volatility and uncertainty of clean energy power generation, resulting in low dispatching efficiency and poor power supply stability in multi-energy power grids.
[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0016] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.
[0017] Example 1, as Figure 1 As shown, this application provides an adaptive scheduling method for multi-energy power grids, the method comprising:
[0018] P10: After performing time-series environmental data collection, establish a prediction time-series environmental dataset based on the time-series environmental data collection results.
[0019] Furthermore, in the execution of timing environment data acquisition, step P10 of this embodiment also includes:
[0020] P11a: Perform time-series acquisition of meteorological data to establish a first-dimensional dataset; P12a: Perform time-series acquisition of environmental perception data to establish a second-dimensional dataset; P13a: Establish the time-series environmental data acquisition results based on the first-dimensional dataset and the second-dimensional dataset.
[0021] It should be understood that a time-series environmental dataset for energy generation forecasting is established by collecting and integrating time-series environmental data.
[0022] Specifically, meteorological data is first collected in a time series to construct the first-dimensional dataset. Meteorological data, such as wind speed, wind direction, sunlight intensity, and temperature, is a key factor affecting energy generation. This data is collected by meteorological monitoring equipment installed within the power grid coverage area and recorded at fixed time intervals to form the first-dimensional dataset. This dataset provides basic meteorological condition information for subsequent energy generation forecasting. For example, as shown in Table 1:
[0023] Table 1: Time-series meteorological data
[0024] Timestamp Wind speed (m / s) Wind direction (°) Light intensity (W / m²) Temperature (°C) humidity(%) 2025 / 5 / 28 0:00 3.2 120 0 18 85 2025 / 5 / 28 1:00 3.5 115 0 17 86 2025 / 5 / 28 2:00 3 110 0 16 87 2025 / 5 / 28 3:00 2.8 105 0 15 88 2025 / 5 / 28 4:00 2.5 100 0 14 89 2025 / 5 / 28 5:00 2.2 95 100 13 90 2025 / 5 / 28 6:00 2 90 200 12 91 2025 / 5 / 28 7:00 1.8 85 300 11 92 2025 / 5 / 28 8:00 1.5 80 400 10 93 2025 / 5 / 28 9:00 1.2 75 500 9 94
[0025] Next, time-series data collection of environmental sensing data is performed to construct the second-dimensional dataset. Environmental sensing data includes, but is not limited to, operational status data of power grid equipment (such as equipment temperature and vibration), power grid topology change data (such as line switching and equipment failure), and other environmental factors related to energy generation (such as the impact of water level changes on hydropower generation). This data is collected by sensors installed on power grid equipment and recorded in a time-series manner. Through the collection of this data, a comprehensive understanding of changes in the power grid operating environment can be obtained, thus providing a more comprehensive reference for energy generation forecasting. The collected environmental sensing data is organized and stored to form the second-dimensional dataset. An example is shown in Table 2:
[0026] Table 2: Temporal Environment Perception Data Table
[0027] Timestamp Equipment temperature (°C) Equipment vibration (mm / s) Power grid frequency (Hz) Water level (m) Line current (A) 2025 / 5 / 28 0:00 45 0.5 50 100 200 2025 / 5 / 28 1:00 46 0.6 50.1 101 205 2025 / 5 / 28 2:00 47 0.7 50.2 102 210 2025 / 5 / 28 3:00 48 0.8 50.3 103 215 2025 / 5 / 28 4:00 49 0.9 50.4 104 220 2025 / 5 / 28 5:00 50 1 50.5 105 225 2025 / 5 / 28 6:00 51 1.1 50.6 106 230 2025 / 5 / 28 7:00 52 1.2 50.7 107 235 2025 / 5 / 28 8:00 53 1.3 50.8 108 240 2025 / 5 / 28 9:00 54 1.4 50.9 109 245
[0028] In this step, time-series data acquisition uses various devices and sensors to periodically acquire power grid and environmental sensing data, forming an independent data dimension that can be compared with meteorological data.
[0029] After completing the two sub-steps described above, the data is integrated and analyzed based on the first and second dimension datasets to establish the time-series environmental data collection results. This step integrates meteorological data and environmental sensing data to create a comprehensive time-series environmental dataset. By combining these two datasets, a multi-dimensional time-series dataset can be formed. This dataset not only covers changes in meteorological conditions but also includes information such as grid load and demand, enabling it to more accurately reflect changes in external conditions for energy generation and providing more comprehensive and accurate data support for subsequent energy generation forecasting.
[0030] In this process, time-series data acquisition refers to the continuous collection of data in chronological order to obtain the patterns of data change over time. This acquisition method can capture the dynamic changes of environmental factors, providing richer information for the prediction model. The first-dimensional dataset and the second-dimensional dataset represent collections of meteorological data and environmental sensing data, respectively, reflecting the external conditions of energy generation from different perspectives. Through comprehensive analysis of these two dimensions of data, a more comprehensive understanding of the operating environment of energy generation can be achieved, thereby improving the accuracy and reliability of energy generation prediction.
[0031] Furthermore, based on the time-series environmental data acquisition results, a prediction time-series environmental dataset is established. Step P10 in this embodiment further includes:
[0032] P11: After performing adaptive data filtering and repair on the time-series environmental data acquisition results, perform dynamic data smoothing processing; P12: Perform backtracking prediction at a multi-level time granularity based on the dynamic data smoothing processing results, perform prediction fusion based on the backtracking prediction results at a multi-level time granularity, and establish a prediction time-series environmental dataset.
[0033] Optionally, the process of establishing the time-series environmental dataset for prediction can be further refined. After completing the collection of time-series environmental data and generating the collection results, this step introduces a data processing and prediction fusion process to improve data quality and prediction accuracy.
[0034] First, adaptive data filtering and repair are performed on the time-series environmental data acquisition results. During actual acquisition, due to sensor malfunctions, environmental interference, or other factors, the acquired data may contain noise, missing values, or outliers. To ensure data accuracy and usability, adaptive data filtering technology is employed to automatically identify and remove noise and outliers based on the statistical characteristics and variation patterns of the data. Simultaneously, missing data points are filled in using interpolation or other repair algorithms to restore data integrity. This process effectively improves data quality, providing a reliable data foundation for subsequent processing and prediction.
[0035] After data filtering and repair, dynamic data smoothing is further performed. The purpose of dynamic data smoothing is to reduce short-term fluctuations and randomness in the data, and to extract long-term trends and patterns. By employing moving averages, exponential smoothing, or other dynamic smoothing algorithms, time-series environmental data is processed to make it smoother, thereby reducing the impact of short-term fluctuations on prediction results. Dynamic data smoothing enhances data stability and consistency, providing more stable data input for subsequent predictive analysis.
[0036] Next, based on the results of dynamic data smoothing, multi-level time-granularity backtesting is performed. Multi-level time-granularity backtesting refers to predictive analysis of historical data at different time scales (such as hours, days, weeks, etc.). By building predictive models at different time granularities, the patterns and trends of data changes at different time scales can be captured. For example, at the hourly time granularity, short-term fluctuations can be reflected more accurately; while at the weekly time granularity, long-term trends can be better grasped. By analyzing the prediction results at different time granularities, a more comprehensive understanding of the data's changing characteristics can be achieved.
[0037] Finally, prediction fusion is performed based on the backtracking prediction results at multiple time granularities to establish a predicted time-series environment dataset. Prediction fusion is the process of comprehensively analyzing and integrating prediction results at different time granularities. By employing weighted averaging, machine learning fusion algorithms, or other fusion methods, prediction results at different time granularities are organically combined to form a comprehensive predicted time-series environment dataset. This process effectively combines the advantages of multiple prediction models, thereby reducing the bias or error that may be generated by a single model prediction. The fused prediction results are more stable and can more accurately reflect future environmental change trends.
[0038] P20: Using the predicted time series environment dataset, energy generation predictions are made for multi-energy power grids, and energy generation prediction results are established. The energy generation prediction results are set with a confidence level indicator.
[0039] Specifically, the energy generation of multi-energy power grids is predicted using a time-series environmental dataset, and corresponding energy generation prediction results are established. Meanwhile, to ensure the reliability and usability of the prediction results, a confidence level indicator is set for the prediction results.
[0040] Specifically, after establishing the predicted time-series environmental dataset, which contains processed and fused high-quality time-series environmental data, reliable data support is provided for energy generation prediction. First, this dataset is used to predict the generation of various energy sources in a multi-energy power grid. Multi-energy power grids typically include multiple energy types, such as wind, solar, hydro, and fossil fuels, each with its own power generation characteristics significantly influenced by environmental factors. For example, the output power of wind power is directly affected by meteorological conditions such as wind speed and direction; the output power of photovoltaic power is closely related to sunlight intensity and temperature. Therefore, based on the predicted time-series environmental dataset and combined with characteristic models of various energy generation equipment (such as the power curve of wind turbines and the photoelectric conversion efficiency model of solar panels), the power generation of different energy types can be predicted, generating energy generation prediction results.
[0041] To assess the reliability of prediction results, this application introduces a confidence level indicator into the energy generation prediction results. The confidence level indicator is a quantitative metric calculated based on the prediction model's confidence interval, historical prediction accuracy, data quality, and other relevant factors, used to measure the credibility of the prediction results. For example, if the prediction model has a high accuracy rate on historical data and the current data quality is good, the confidence level indicator of the prediction results will be high; conversely, if the prediction model has significant uncertainty or the data contains a lot of noise, the confidence level indicator will be lower. By setting a confidence level indicator, dispatchers can more intuitively understand the reliability of the prediction results, thereby making reasonable use of this prediction information in subsequent dispatch decisions.
[0042] Therefore, by using a time-series environmental dataset to predict energy generation in a multi-energy grid and setting a confidence level label for the prediction results, we can not only provide prediction information for future energy generation, but also quantify the reliability of the prediction results through the confidence level label.
[0043] P30: After decomposing the power demand into N sub-power demand, each sub-power demand is used as a game agent to perform a time-adaptive game with the energy generation forecast results.
[0044] Furthermore, step P30 in this embodiment of the application also includes:
[0045] P31: Configure the power supply stability target and task adaptation target of the game agent according to the sub-power supply demand; P32: Perform independent adaptation analysis between the game agent and the energy generation prediction results based on the power supply stability target and task adaptation target, and establish an independent adaptation value; P33: Establish a time-series stability threshold using the trust level identifier; P34: Perform group time-series adaptation game based on the independent adaptation value and the time-series stability threshold to generate the time-series adaptation game result.
[0046] It should be understood that this involves decomposing power demand into multiple sub-demands and managing each sub-demand as an independent game agent. In this way, each sub-demand in the power grid can act as an independent game agent, engaging in time-series adaptation games with energy generation forecasts. The goal of this process is to optimize the matching between power demand and generation forecasts, ensuring the efficient operation of the power grid.
[0047] Specifically, the power supply stability objective and task adaptation objective of the game agent are first configured based on the sub-power supply demand. Each sub-power supply demand plays an independent role in the game agent and must satisfy certain specific objectives. During configuration, the power supply stability objective for each game agent is first determined, which ensures the stability of power supply demand and avoids negative impacts on the power grid caused by fluctuations in power supply. In addition, a task adaptation objective needs to be configured, which aims to optimize the matching between sub-demand scheduling and generation forecasting to ensure that the power grid meets demand while improving energy utilization and overall system efficiency.
[0048] Next, based on the power supply stability objective and the task adaptation objective, an independent adaptation analysis is conducted between the game agent and the energy generation prediction results to establish an independent adaptation value. In this process, each game agent independently analyzes and matches itself with the energy generation prediction results according to its configured power supply stability objective and task adaptation objective. By quantifying the adaptation degree between each game agent and the generation prediction results, an independent adaptation value is obtained. This value reflects whether the game agent can meet its power supply demand objective at that moment and the degree of adaptation with the prediction results. A higher independent adaptation value indicates a better match between the agent and the generation prediction results.
[0049] Simultaneously, a time-series stability threshold is established using the confidence level indicators from energy generation forecasts. The confidence level indicators reflect the reliability of the forecast results and thus serve as a key indicator for assessing power supply stability. The setting of the time-series stability threshold is based on the statistical distribution of the confidence level indicators and the analysis of historical data. For example, if the confidence level indicator is high within a certain time period, a higher time-series stability threshold can be set, indicating higher power supply stability during that time period; conversely, if the confidence level indicator is low, a lower time-series stability threshold needs to be set to remind the dispatch system to take additional stabilization measures during that time period.
[0050] Finally, based on the independent fit value and the time-series stability threshold, a group time-series adaptation game is executed. The group time-series adaptation game is an optimization process in a multi-agent system, aiming to achieve the optimal match between the overall power demand and the energy generation prediction results while satisfying the power supply stability and task adaptation objectives of all sub-power demands. Various algorithms can be employed in the game process, such as Nash equilibrium algorithms and multi-objective optimization algorithms. Through the group time-series adaptation game, the final time-series adaptation game result is generated, which includes the optimal power supply scheme for each sub-power demand, the corresponding energy generation type, the power supply time window, and necessary compensation measures.
[0051] In summary, by decomposing power demand into multiple sub-demands and treating them as independent adaptive agents in a game theory approach, efficient matching of power demand with energy generation forecasts is achieved. Furthermore, by configuring power supply stability objectives, task adaptation objectives, and combining trust level identifiers and time-series stability thresholds, each decision in the scheduling process is optimized, ultimately generating a stable and efficient time-series adaptation game result, thereby ensuring the efficient and stable operation of the power grid.
[0052] Furthermore, step P34 in this embodiment of the application also includes:
[0053] P34-1: Sort the game agents sequentially based on the independent adaptation value; P34-2: Use the temporal stability threshold to screen the group adaptation of the sequential sorting results, and establish retained adapted game agents, eliminated game agents, and temporarily stored game agents; P34-3: Under the condition of supply and demand scheduling of retained adapted game agents, execute the reconstruction adaptation game of temporarily stored game agents, the reconstruction adaptation game includes executing supply and demand interruption penalty using a disconnection penalty mechanism; P34-4: Complete the group temporal adaptation game based on the reconstruction adaptation game results.
[0054] Optionally, the process of group time-series adaptation game can be further refined. By sorting the game agents, adapting and filtering, reconstructing the adaptation game, and generating the final game result, the matching between power supply demand and energy generation forecast results can be effectively optimized.
[0055] When performing group time-series adaptation game theory, all game agents are first ranked based on their Independent Fitness Value (IQV). IQV reflects the degree of matching between the game agent and the energy generation forecast; a higher value indicates a better match. By ranking the game agents, it is possible to identify which agents best meet the scheduling requirements, thus prioritizing the fulfillment of these agents' needs. This step helps determine which agents should participate in scheduling first and which can temporarily refrain from participation, thereby achieving a rational allocation of scheduling resources.
[0056] Subsequently, a time-series stability threshold was used to screen the sequential ranking results for group suitability. This threshold, based on a trust level indicator of energy generation forecasts, is used to assess power supply stability; agents exceeding the threshold indicate relatively stable demand and are suitable for inclusion in the final scheduling scheme. Through screening, the game agents were categorized into three types: retained suitable game agents, eliminated game agents, and temporarily stored game agents. Retained suitable game agents are those whose suitability and stability both meet the requirements and can directly enter the supply and demand scheduling process; eliminated game agents are those whose suitability or stability does not meet the requirements and cannot participate in the current scheduling; temporarily stored game agents are in an intermediate state and require further evaluation and adjustment.
[0057] Under the condition of retaining the supply and demand scheduling of the temporarily stored game agents, a reconfiguration adaptation game is performed. The purpose of the reconfiguration adaptation game is to reassess the adaptability of the temporarily stored game agents and punish supply and demand interruptions through a disconnection penalty mechanism. In this process, the supply and demand matching of the temporarily stored game agents is first checked to ensure that their demands can be reasonably met. If their demands cannot be met under the current power generation forecast, the system will execute the disconnection penalty mechanism, that is, to impose a penalty on the agents whose supply and demand are interrupted, so as to promote the agents to re-adapt. In this way, the temporarily stored game agents can make necessary adjustments to optimize their supply and demand matching so as to re-enter the scheduling queue.
[0058] After reconstructing the adaptation game, a new set of game results will be obtained. Based on these reconstructed adaptation game results, the group time-series adaptation game is finally completed. The goal of this step is to summarize the adjustment and optimization results of all game agents to generate the final time-series adaptation game result, ensuring that the power supply demand of the entire power grid can be optimally matched with the power generation forecast results. This result not only includes the optimal power supply scheme for each sub-power supply demand, the corresponding energy generation type, and the power supply time window, but also ensures the stability of power supply through a disconnection penalty mechanism.
[0059] P40: Configure the sub-power supply demand scheduling of clean energy according to the result of the time-series adaptation game, and establish compensation scheduling for the unconfigured sub-power supply demand.
[0060] Specifically, based on the previously generated time-series adaptation game results, the power supply demand of sub-energy grids is scheduled and configured. This step aims to optimize the utilization of clean energy and ensure stable power supply from the grid.
[0061] Specifically, the first step is to schedule and configure the sub-supply demands of clean energy based on the results of the time-series adaptation game. The results of the time-series adaptation game provide the optimal matching scheme between each sub-supply demand and the energy generation forecast, including information such as the corresponding energy generation type, supply time window, and degree of adaptation. During the configuration process, clean energy resources are prioritized for scheduling to meet the matched sub-supply demands, ensuring that, under reasonable power generation forecasts, renewable energy sources (such as solar and wind power) are used as much as possible to meet grid demand.
[0062] For each sub-supply demand, based on its matching result in the time-series adaptation game, it is determined which clean energy source will supply power during a specific time period, and the corresponding power output is allocated. For example, if a sub-supply demand highly matches the predicted output power of wind power generation during a certain time period, then that sub-supply demand will be allocated to wind power generation. In this way, the power generation capacity of clean energy sources can be maximized, dependence on traditional fossil fuels can be reduced, and the operating efficiency and environmental benefits of the power grid can be improved.
[0063] After completing the dispatch of sub-supply demands from clean energy sources, some sub-supply demands may not be fully met by clean energy. This could be due to limited clean energy generation capacity, forecasting errors, or other uncontrollable factors. To ensure stable power supply to the grid, compensatory dispatch is needed for the unmet sub-supply demands. The purpose of compensatory dispatch is to meet the remaining power demands using other energy types (such as traditional fossil fuel power generation, energy storage systems, etc.). Specifically, appropriate compensatory energy types can be selected based on the actual situation of the grid and dispatch strategies. For example, if the grid has energy storage systems (such as battery storage, pumped hydro storage, etc.), the charging and discharging operations of these systems can be prioritized to meet the unmet sub-supply demands. Energy storage systems have the characteristics of rapid response and flexible adjustment, enabling them to fill power gaps in a short time and improve the stability and reliability of the grid. If the energy storage system cannot meet all the demands, or if the energy storage system itself also needs to replenish energy, the output of traditional fossil fuel power generation equipment can be adjusted to ensure that the total power supply capacity of the grid meets the demand.
[0064] When establishing compensatory dispatch, the economic efficiency and flexibility of the dispatch must also be considered. For example, optimization algorithms can be used to determine the optimal combination of compensatory energy sources and dispatch schemes to minimize the cost of compensatory dispatch while ensuring the safe and stable operation of the power grid. Furthermore, the establishment of compensatory dispatch also needs to be coordinated with the sub-dispatch of clean energy demand to achieve optimized dispatch of the entire power grid.
[0065] This configuration ensures the maximum utilization of clean energy while supplementing the dispatch of surplus electricity demand, thereby optimizing energy efficiency and maintaining the stable operation of the power grid.
[0066] P50: Adaptive scheduling management is performed based on the sub-power supply demand scheduling and the compensation scheduling.
[0067] Furthermore, step P50 in this embodiment of the application also includes:
[0068] P51: Perform real-time updates of environmental data and reconstruct energy generation forecast results using the real-time update results; P52: Perform transfer gain analysis of scheduling transfer based on the reconstructed energy generation forecast results and the compensation scheduling, and establish transfer gain analysis results; P53: Obtain the power supply maintenance threshold. If the transfer gain analysis results exceed the power supply maintenance threshold, the corresponding compensation scheduling is converted into clean energy sub-power supply demand scheduling for adaptive scheduling management.
[0069] Optionally, adaptive scheduling management can be implemented based on the configuration of sub-power supply demand scheduling and compensation scheduling to ensure the stable operation of the power grid and the optimal utilization of energy resources.
[0070] First, environmental data is updated in real time, and the results are used to reconstruct energy generation forecasts. The operating status of the power grid and environmental factors (such as meteorological data and load changes) are constantly changing, necessitating real-time updates to the collected environmental data. This real-time data effectively reflects current power generation conditions, load demand, and changes in the meteorological environment. Based on this real-time updated data, the original energy generation forecasts are reconstructed, making the forecasts more closely reflect the current situation and improving accuracy. This reconstruction process is crucial for ensuring that adaptive dispatch can flexibly respond to real-time changes. For example, if real-time meteorological data shows a sudden increase in wind speed, the predicted output power of wind power will be adjusted upwards accordingly; conversely, if sunlight intensity decreases, the predicted output power of photovoltaic power will need to be adjusted downwards. By updating and reconstructing forecasts in real time, dispatch decisions are ensured to always be based on the latest information.
[0071] Subsequently, based on the reconstructed energy generation forecasts and existing compensatory scheduling, a transfer gain analysis of the dispatch shift is conducted. The purpose of the transfer gain analysis is to assess the potential benefits of converting compensatory scheduling into clean energy sub-demand scheduling. Specifically, it analyzes whether, under the current energy generation forecast conditions, adjusting the dispatch scheme can shift the power supply tasks originally undertaken by traditional energy or energy storage systems to clean energy generation, thereby achieving higher economic benefits, environmental benefits, or system stability. For example, if the reconstructed forecasts show that there is surplus clean energy generation capacity in a certain period, and shifting the power supply tasks can reduce the charging and discharging losses of energy storage systems or reduce the cost of using traditional energy, then the dispatch shift has a positive gain. By establishing the transfer gain analysis results, a quantitative basis is provided for subsequent dispatch adjustments.
[0072] Finally, the power supply retention threshold is obtained, and scheduling decisions are made based on the transfer gain analysis results. The power supply retention threshold is a key parameter pre-set by the system to measure the feasibility and necessity of scheduling transfers. This threshold comprehensively considers multiple factors such as grid operation stability, economic efficiency, and clean energy utilization rate. If the transfer gain analysis results indicate that the potential benefits of scheduling transfers exceed the power supply retention threshold, the corresponding compensatory scheduling is converted into clean energy sub-power supply demand scheduling. This conversion process not only optimizes the utilization efficiency of clean energy but also improves the overall operating performance of the grid. By dynamically adjusting the scheduling scheme, it can be ensured that the grid maintains its optimal state during real-time operation, achieving adaptive scheduling management.
[0073] Furthermore, the embodiments of this application also include step P53a:
[0074] If the transfer gain analysis result fails to exceed the power supply maintenance threshold, then the sub-power supply demand scheduling and the compensation scheduling will be maintained under adaptive scheduling management.
[0075] If the transfer gain analysis results do not exceed the power supply maintenance threshold, it indicates that the current compensation scheduling and sub-power supply demand scheduling are sufficient to meet the grid's needs, and the system has not experienced excessive fluctuations or instability. In this case, the system will continue to maintain the existing scheduling scheme without any additional adjustments.
[0076] Specifically, the transfer gain analysis results show that the adjustment range of the compensation scheduling meets the power grid's power supply stability requirements and does not exceed the preset threshold. Therefore, the system does not need to make further changes to power supply scheduling and can continue to execute the current sub-power supply demand scheduling and compensation scheduling scheme. At this time, the system will continue to perform adaptive scheduling management using the existing scheduling strategy to ensure the continuous satisfaction of power supply demand, while avoiding unnecessary scheduling adjustments, reducing resource waste and system burden. In this way, the system can avoid unnecessary adjustments under stable operating conditions, thereby improving the efficiency of energy dispatching and the stability of the system.
[0077] Furthermore, based on the sub-power supply demand scheduling and the compensation scheduling, adaptive scheduling management is performed. In this embodiment, step P50 further includes:
[0078] P51b: Monitor the actual power supply status of the power supply demand and establish monitoring feedback; P52b: Use the monitoring feedback to perform adaptive scheduling optimization of the sub-power supply demand scheduling and the compensation scheduling.
[0079] Specifically, this application's embodiments introduce actual power supply status monitoring of power demand in adaptive scheduling management. By deploying intelligent monitoring devices in the power grid, the actual power supply status of each sub-power demand is monitored in real time, generating monitoring feedback, and the monitoring feedback is used for scheduling optimization.
[0080] First, the actual power supply status is monitored to assess power demand, and a monitoring feedback mechanism is established. This step involves deploying sensors and monitoring equipment to monitor the power grid's power supply in real time. Monitoring includes real-time data on power load, power generation, and demand response, ensuring timely feedback on the grid's operational status. Through this monitoring, the system can identify the discrepancy between the actual power supply status and the expected dispatch. If insufficient power supply or overload is detected during actual power supply, the system will capture these anomalies in real time, establish a feedback mechanism, and use this actual data as input information for subsequent dispatch optimization.
[0081] After obtaining monitoring feedback, this feedback information can be used to adaptively optimize sub-supply demand scheduling and compensation scheduling. For example, if the monitoring feedback shows that the actual power supply for a certain sub-supply demand is lower than the predicted value, and there is a power supply gap, the scheduling system can dynamically adjust the compensation scheduling to increase the power supply during the corresponding period. Conversely, if the actual power supply for a certain sub-supply demand exceeds the predicted value, the power supply during that period can be appropriately reduced to optimize energy allocation. In addition, monitoring feedback can also be used to adjust the scheduling priority of clean energy. For example, when the clean energy generation capacity exceeds expectations, high-priority sub-supply demands can be prioritized to further improve the utilization rate of clean energy.
[0082] By introducing actual power supply status monitoring of power demand and scheduling optimization based on monitoring feedback, the adaptive scheduling management of this application embodiment can not only adjust the scheduling scheme according to real-time environmental data, but also dynamically optimize scheduling decisions according to the actual power supply status. This dual optimization mechanism ensures the real-time performance, accuracy, and flexibility of the scheduling scheme, effectively copes with the complex situations in the operation of multi-energy power grids, and further improves the operating efficiency, stability, and reliability of the power grid.
[0083] In summary, the embodiments of this application have at least the following technical effects:
[0084] This application establishes a time-series environmental dataset through the collection and prediction of such data. Using this dataset, it predicts the power generation of a multi-energy power grid and generates power generation prediction results with trust level indicators. Power demand is decomposed into multiple sub-demands, and each sub-demand is used as a game agent to engage in time-series adaptation game theory with the power generation prediction results. This allows for the configuration of clean energy dispatch, with any unconfigured demand being supplemented through compensatory dispatch.
[0085] The technology achieves the goal of improving the scheduling efficiency and power supply stability of the energy grid by using time-series environmental data prediction and game-theoretic intelligent agent scheduling to realize efficient coordinated scheduling of clean energy and traditional energy in multi-energy power grids.
[0086] Example 2, based on the same inventive concept as the adaptive dispatch method for multi-energy power grids in the foregoing examples, such as... Figure 2 As shown, this application provides an adaptive dispatch system for multi-energy power grids. The system and method embodiments in this application are based on the same inventive concept. The system includes:
[0087] The time-series environmental data prediction module 11 is used to establish a predicted time-series environmental dataset based on the time-series environmental data acquisition results after performing time-series environmental data acquisition.
[0088] Energy generation prediction module 12 is used to predict energy generation of multi-energy power grids using the prediction time series environmental dataset, and to establish energy generation prediction results. The energy generation prediction results are set with a trust level indicator.
[0089] The timing adaptation game module 13 is used to decompose the power supply demand into N sub-power supply demands, and then use each sub-power supply demand as a game agent to perform a timing adaptation game with the energy generation prediction results.
[0090] The scheduling demand configuration module 14 is used to configure the scheduling of sub-power supply demand of clean energy according to the result of time-series adaptation game, and to establish compensation scheduling for the unconfigured sub-power supply demand.
[0091] The adaptive scheduling management module 15 is used to perform adaptive scheduling management based on the sub-power supply demand scheduling and the compensation scheduling.
[0092] Furthermore, the time-series environmental data prediction module 11 is also used to perform the following steps:
[0093] Meteorological data is collected in a time series to establish a first-dimensional dataset; environmental sensing data is collected in a time series to establish a second-dimensional dataset; and the time series environmental data collection results are established based on the first-dimensional dataset and the second-dimensional dataset.
[0094] Furthermore, the time-series environmental data prediction module 11 is also used to perform the following steps:
[0095] After performing adaptive data filtering and repair on the time-series environmental data acquisition results, dynamic data smoothing is performed; based on the results of dynamic data smoothing, multi-level time-granularity backtracking prediction is performed, and prediction fusion is performed based on the results of multi-level time-granularity backtracking prediction to establish a prediction time-series environmental dataset.
[0096] Furthermore, the timing-adaptive game module 13 is also used to perform the following steps:
[0097] Configure the power supply stability target and task adaptation target of the game agent according to the sub-power supply demand; perform independent adaptation analysis between the game agent and the energy generation prediction results based on the power supply stability target and task adaptation target, and establish independent adaptation value; establish time-series stability threshold using the trust degree identifier; perform group time-series adaptation game based on the independent adaptation value and the time-series stability threshold to generate time-series adaptation game result.
[0098] Furthermore, the timing-adaptive game module 13 is also used to perform the following steps:
[0099] The game agents are ordered based on the independent adaptation values; the group adaptation of the ordered results is screened using the temporal stability threshold to establish retained adapted game agents, eliminated game agents, and temporarily stored game agents; under the condition of supply and demand scheduling of retained adapted game agents, a reconstruction adaptation game is performed on the temporarily stored game agents, the reconstruction adaptation game including the use of a disconnection penalty mechanism to execute supply and demand interruption penalties; the group temporal adaptation game is completed based on the reconstruction adaptation game results.
[0100] Furthermore, the adaptive scheduling management module 15 is also used to perform the following steps:
[0101] The system performs real-time updates of environmental data and reconstructs energy generation forecasts using the real-time update results. Based on the reconstructed energy generation forecasts and the compensation scheduling, it performs transfer gain analysis of scheduling shifts and establishes transfer gain analysis results. It obtains a power supply maintenance threshold. If the transfer gain analysis results exceed the power supply maintenance threshold, the corresponding compensation scheduling is converted into a clean energy sub-power supply demand scheduling for adaptive scheduling management.
[0102] Furthermore, the adaptive scheduling management module 15 is also used to perform the following steps:
[0103] If the transfer gain analysis result fails to exceed the power supply maintenance threshold, then the sub-power supply demand scheduling and the compensation scheduling will be maintained under adaptive scheduling management.
[0104] Furthermore, the adaptive scheduling management module 15 is also used to perform the following steps:
[0105] The actual power supply status is monitored to determine the power supply demand, and a monitoring feedback is established. The monitoring feedback is then used to perform adaptive scheduling optimization for the sub-power supply demand scheduling and the compensation scheduling.
[0106] Example 3, Exemplary Electronic Device, as described below. Figure 3 The present application describes the electronic device according to its embodiments.
[0107] Based on the same inventive concept as the adaptive scheduling method for multi-energy power grids in the foregoing embodiments, this application also provides an adaptive scheduling system for multi-energy power grids, including: a processor coupled to a memory for storing a program, wherein when the program is executed by the processor, the system performs the steps of the method described in Embodiment 1.
[0108] The electronic device 300 includes a processor 302, a communication interface 303, and a memory 301. Optionally, the electronic device 300 may also include a bus architecture 304. The communication interface 303, processor 302, and memory 301 can be interconnected via the bus architecture 304; the bus architecture 304 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus architecture 304 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0109] Processor 302 may be a CPU, microprocessor, ASIC, or one or more integrated circuits used to control the execution of programs according to the present application.
[0110] Communication interface 303 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), wired access network, etc.
[0111] Memory 301 can be ROM or other types of static storage devices capable of storing static information and instructions, RAM or other types of dynamic storage devices capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory can exist independently and be connected to the processor via bus architecture 304. Memory can also be integrated with the processor.
[0112] The memory 301 stores computer execution instructions for implementing the scheme of this application, and the processor 302 controls the execution. The processor 302 executes the computer execution instructions stored in the memory 301, thereby realizing the adaptive scheduling method for multi-energy power grids provided in the above embodiments of this application.
[0113] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.
[0114] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
[0115] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.
Claims
1. An adaptive dispatching method for multi-energy power grids, characterized in that, The method includes: After performing time-series environmental data collection, a prediction time-series environmental dataset is established based on the time-series environmental data collection results; Using the aforementioned time-series environmental dataset, energy generation predictions are made for multi-energy power grids, and energy generation prediction results are established, with a trust level indicator set for the energy generation prediction results. After decomposing the power demand into N sub-power demand, each sub-power demand is treated as a game agent and a time-adaptive game is played with the energy generation forecast results. Configure the sub-power supply demand scheduling of clean energy according to the results of the time-series adaptation game, and establish compensation scheduling for the unconfigured sub-power supply demand. Adaptive scheduling management is performed based on the sub-power supply demand scheduling and the compensation scheduling.
2. The adaptive scheduling method for multi-energy power grids as described in claim 1, characterized in that, The process of treating each sub-power supply demand as a game agent and performing a time-adaptive game with the energy generation forecast results includes: Configure the power supply stability target and task adaptation target of the game intelligence agent according to the sub-power supply requirements; Based on the power supply stability target and task adaptation target, an independent adaptation analysis is performed between the game agent and the energy generation prediction results to establish an independent adaptation value. Establish a time-series stability threshold using the aforementioned trust level identifier; Based on the independent adaptation value and the temporal stability threshold, a group temporal adaptation game is performed to generate the temporal adaptation game result.
3. The adaptive scheduling method for multi-energy power grids as described in claim 2, characterized in that, The process of performing group time-series adaptation game based on the independent fit value and the time-series stability threshold to generate the time-series adaptation game result includes: The order of the game agents is determined based on the independent adaptation values. Using the time-series stability threshold, the group adaptation screening of the sequential sorting results is carried out to establish retained adapted game agents, eliminated game agents, and temporarily stored game agents; While preserving the supply and demand scheduling of the adaptive game agents, a reconstructed adaptive game of the temporarily stored game agents is performed, wherein the reconstructed adaptive game includes implementing a supply and demand interruption penalty using a disconnection penalty mechanism; Complete the group temporal adaptation game based on the results of the reconstruction adaptation game.
4. The adaptive scheduling method for multi-energy power grids as described in claim 1, characterized in that, The execution timing environment data acquisition includes: Meteorological data were collected in a time series to establish the first-dimensional dataset; Perform time-series acquisition of environmental perception data to establish a second-dimensional dataset; The time-series environmental data acquisition results are established based on the first dimension dataset and the second dimension dataset.
5. The adaptive dispatching method for multi-energy power grids as described in claim 4, characterized in that, The step of establishing a predictive time-series environment dataset based on the time-series environment data collection results includes: After performing adaptive data filtering and repair on the time-series environmental data acquisition results, dynamic data smoothing processing is performed. Based on the results of dynamic data smoothing, backtracking predictions at multiple time granularities are performed. Based on the backtracking prediction results at multiple time granularities, prediction fusion is performed to establish a prediction time series environment dataset.
6. The adaptive scheduling method for multi-energy power grids as described in claim 1, characterized in that, The adaptive scheduling management based on the sub-power supply demand scheduling and the compensation scheduling includes: Perform real-time updates of environmental data and use the real-time update results to reconstruct energy generation forecasts; Based on the reconstructed energy generation forecast results and the compensation scheduling, a transfer gain analysis of the scheduling transfer is performed, and a transfer gain analysis result is established. If the transfer gain analysis result exceeds the power supply retention threshold, the corresponding compensation scheduling is converted into clean energy sub-power supply demand scheduling for adaptive scheduling management.
7. The adaptive scheduling method for multi-energy power grids as described in claim 6, characterized in that, If the transfer gain analysis result fails to exceed the power supply maintenance threshold, then the sub-power supply demand scheduling and the compensation scheduling will be maintained under adaptive scheduling management.
8. The adaptive scheduling method for multi-energy power grids as described in claim 1, characterized in that, The adaptive scheduling management based on the sub-power supply demand scheduling and the compensation scheduling also includes: Monitor the actual power supply status to meet power demand and establish a monitoring feedback mechanism; The monitoring feedback is used to perform adaptive scheduling optimization for the sub-power supply demand scheduling and the compensation scheduling.
9. An adaptive dispatch system for multi-energy power grids, characterized in that, The system includes: A time-series environmental data prediction module is used to establish a predicted time-series environmental dataset based on the time-series environmental data acquisition results after performing time-series environmental data acquisition. An energy generation prediction module is used to predict energy generation in a multi-energy grid using the prediction time series environmental dataset, and to establish energy generation prediction results. The energy generation prediction results are equipped with a trust level indicator. The timing adaptation game module is used to decompose the power supply demand into N sub-power supply demands, and then use each sub-power supply demand as a game agent to perform a timing adaptation game with the energy generation prediction results. The scheduling demand configuration module is used to configure the sub-power supply demand scheduling of clean energy according to the result of the time-series adaptation game, and to establish compensation scheduling for the unconfigured sub-power supply demand. An adaptive scheduling management module is used to perform adaptive scheduling management based on the sub-power supply demand scheduling and the compensation scheduling.
10. An electronic device, characterized in that, include: A processor coupled to a memory for storing a program that, when executed by the processor, causes the system to perform the steps of the method as claimed in any one of claims 1 to 8.