An interactive control method and system for multi-scenario distribution network

Abstract

The present invention provides an interactive control method and system for a multi-scenario distribution network, which relates to the technical field of power grid control. The method includes: analyzing the current power consumption data of each user group under all scenario nodes according to the real-time power distribution data of all scenario nodes in the distribution network, and obtaining the future power consumption data of the user group in a preset future time period; formulating a sub-control decision for each user group based on the future power consumption data and the control target of each user group; determining the fitness of each sub-control decision for the final control decision of the distribution network according to the sub-control decision of each user group; performing multi-objective optimization according to the fitness, obtaining the final control decision, and controlling the distribution network to distribute power. The present invention improves the control flexibility and adaptability of the distribution network in multiple scenarios through intelligent data acquisition and analysis, power consumption feature identification, multi-objective strategy formulation, and optimized decision selection.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid control, and in particular to an interactive control method and system for a multi-scenario distribution network. Background Art

[0002] As an important part of the power system, the distribution network must not only meet the basic needs of traditional power supply, but also adapt to the access of emerging loads such as distributed energy, electric vehicle charging stations, and smart homes. These changes have brought about complex needs of multiple user groups in multiple scenarios, and put forward higher requirements on the regulation and control capabilities of the distribution network. There are many user groups in multiple scenarios. The traditional distribution network needs to serve multiple user groups such as industry, commerce, and residents in multiple scenarios. Each user group has its own unique electricity needs and preferences. For example, industrial users may pay more attention to the stability and reliability of power supply, while residential users may pay more attention to the rationality of electricity prices and the convenience of electricity use.

[0003] In existing technologies, distribution network control measures often focus on a single goal, such as minimizing operating costs or maximizing energy efficiency. However, control strategies based on a single goal are difficult to meet the diverse needs of multiple user groups in multiple scenarios. For example, control measures taken to reduce costs will sacrifice the reliability of power supply. In addition, with the large-scale access of renewable energy, distribution networks also need to deal with the intermittency and uncertainty of energy sources such as wind and solar energy, which further increases the complexity of control and causes the distribution network to be unable to meet the electricity needs of different user groups in multiple scenarios. Summary of the invention

[0004] The problem solved by the present invention is how to improve the flexibility and adaptability of distribution network regulation.

[0005] In order to solve the above problems, the present invention provides an interactive control method for a multi-scenario distribution network, comprising:

[0006] Obtain real-time power distribution data of all scene nodes in the distribution network;

[0007] Analyze the real-time power distribution data to obtain current power consumption data of each user group under all the scenario nodes;

[0008] Perform prediction based on the current electricity consumption data to obtain future electricity consumption data of the user group within a preset future time period;

[0009] According to the future electricity consumption data in the preset future time period, the electricity consumption characteristics of each user group under all the scene nodes are obtained, and the electricity consumption characteristics include the electricity consumption mode, peak and valley time periods and load trends of the user group;

[0010] Determine the control target of each user group under all the scenario nodes in the preset future time period according to the power consumption pattern, the peak and valley time periods and the load trend of each user group;

[0011] Inputting the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model to obtain a sub-control decision for each of the user groups;

[0012] Determining, according to the sub-regulation decision of each user group, the adaptability of each sub-regulation decision to the final regulation decision of the distribution network;

[0013] Performing multi-objective optimization according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network;

[0014] The power distribution network is controlled to distribute power according to the final regulation decision.

[0015] Optionally, the analyzing according to the real-time power distribution data to obtain the current power consumption data of each user group under all the scene nodes includes:

[0016] Acquire power consumption indicators of all the scene nodes according to the real-time power distribution data, wherein the power consumption indicators include current power, voltage, current and power of the scene nodes;

[0017] Determine the current power consumption data of the user group according to the power consumption index combined with the power proportion of each of the user groups under the scenario node;

[0018] wherein, according to the voltage and current of the scene node, the total power of the scene node is determined;

[0019] Determining the current power usage of the user group according to the total power and the power proportion;

[0020] The current power usage data of the user group is determined according to the current power usage.

[0021] Optionally, the predicting according to the current power consumption data to obtain future power consumption data of the user group within a preset future time period includes:

[0022] Acquiring influencing factor data within the preset future time period, the influencing factor data including the geographical location and climate data of the user group;

[0023] Determine a corresponding prediction model according to the current power consumption data, the geographical location and the climate data, and input the current power consumption data, the geographical location and the climate data into the prediction model;

[0024] Batch prediction is performed on each of the user groups through the prediction model to obtain the future electricity consumption data of the user group in the preset future time period.

[0025] Optionally, obtaining the power consumption characteristics of each user group under all the scene nodes according to the future power consumption data in the preset future time period includes:

[0026] The user group is grouped according to the future electricity consumption data of the user group in the preset future time period through the objective function of the K-means clustering algorithm, and the user group is divided into a plurality of clusters;

[0027] Place Description The objective function is:

[0028] J= ;

[0029] in, is the number of clusters, is the user group of the i-th cluster, is the center point of the power consumption feature of the i-th cluster, x is the future power consumption data of the user group, and J is the sum of the squares of the distances from all the clusters to the center point of the power consumption feature;

[0030] Feature extraction is performed based on the future electricity usage data of the user group of each cluster to obtain the electricity usage features of the user group.

[0031] Optionally, the extracting features based on the future electricity consumption data of each clustered user group to obtain the electricity consumption features of the user group, the electricity consumption features including the electricity consumption mode, peak and valley time periods, and load trends of the user group, include:

[0032] Obtaining a load curve of the user group in the preset future time period according to the future power consumption data of the clustered user group;

[0033] The electricity consumption behavior of the user group in the preset future time period is predicted according to the load curve to obtain the electricity consumption pattern, the peak and valley time periods and the load trend of the user group in the preset future time period.

[0034] Optionally, determining the control target of each user group in the preset future time period under all the scenario nodes according to the power consumption pattern, the peak and valley time periods, and the load trend of each user group includes:

[0035] According to the power demand of the user group in the preset future time period, obtaining the expected power distribution data in the preset future time period;

[0036] The control target is obtained according to the power consumption pattern of the user group, the peak and valley time periods, and the load trend and the expected power distribution data corresponding to each of the user groups.

[0037] Optionally, inputting the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model to obtain a sub-control decision for each user group includes:

[0038] Inputting the regulation target of the user group and the current power consumption data of the user group into a multi-objective strategy model;

[0039] The current power consumption data and the control target are integrated to obtain an integrated feature vector;

[0040] Assigning a weight to each feature vector in the integrated feature vector, and then solving the multi-objective strategy model through an optimization algorithm to obtain a Pareto optimal solution set;

[0041] Screening is performed according to the Pareto optimal solution set to obtain the sub-regulation decision of the user group.

[0042] Optionally, determining, according to the sub-regulation decision of each user group, the adaptability of each sub-regulation decision to the final regulation decision of the distribution network includes:

[0043] Encoding the sub-regulation decision to obtain a binary code of the sub-regulation decision;

[0044] The adaptability of the sub-regulation decision to the final regulation decision of the power distribution network is determined by a fitness function according to the binary encoding of the sub-regulation decision.

[0045] Optionally, performing multi-objective optimization according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network includes:

[0046] Initializing according to the sub-regulation decision of the user group to obtain an initial population;

[0047] Selecting the sub-regulation decision from the initial population according to the fitness, and iterating the selected sub-regulation decision through a crossover mutation operation to form a new population;

[0048] Updating the fitness of the sub-regulation decision in the new population to obtain the updated fitness;

[0049] When the updated fitness reaches a preset threshold, the Pareto optimal solution set of the final control decision is obtained from the new population formed in the last iteration;

[0050] According to the Pareto optimal solution set, the final control decision is obtained.

[0051] The present invention also provides an interactive control system for a multi-scenario distribution network, comprising:

[0052] A data acquisition unit, used to acquire real-time power distribution data of all scene nodes in the distribution network;

[0053] A first analysis unit, configured to analyze the real-time power distribution data to obtain current power consumption data of each user group under all the scenario nodes;

[0054] A prediction unit, configured to make a prediction based on the current power consumption data to obtain future power consumption data of the user group within a preset future time period;

[0055] A second analysis unit is used to obtain the power consumption characteristics of each user group under all the scene nodes according to the future power consumption data in the preset future time period, wherein the power consumption characteristics include the power consumption mode, peak and valley time periods, and load trends of the user group;

[0056] Determine the control target of each user group under all the scenario nodes in the preset future time period according to the power consumption pattern, the peak and valley time periods and the load trend of each user group;

[0057] A first decision unit, configured to input the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model to obtain a sub-control decision for each of the user groups;

[0058] Determining, according to the sub-regulation decision of each user group, the adaptability of each sub-regulation decision to the final regulation decision of the distribution network;

[0059] A second decision-making unit, configured to perform multi-objective optimization according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network;

[0060] A control unit is used to control the power distribution network to distribute power according to the final control decision.

[0061] The interactive control method and system of the multi-scenario distribution network of the present invention accurately grasps the current power consumption of different user groups by acquiring and analyzing the power consumption data of all scenario nodes in the distribution network in real time. Then, the prediction technology is used to predict the power demand of the user group in the future time period, providing forward-looking information for the control decision. Not only the power demand is considered, but also the power consumption mode, peak and valley time period and load trend of the user are analyzed, so as to customize the personalized power consumption characteristics for each user group. The identification of these characteristics provides a more accurate basis for formulating the control target, so that the control strategy can better meet the specific needs of different user groups. Through the multi-objective strategy model, the control target of each user group is combined with the current power consumption data to generate a sub-control decision. This method not only considers a single goal, but also integrates multiple goals, such as cost, efficiency, reliability, etc., so as to achieve the optimal configuration of resources while meeting diversified needs. Through fitness evaluation and multi-objective optimization, the method can select the optimal final control decision from multiple possible sub-control decisions, ensuring that the distribution network can achieve cost-effectiveness while also ensuring the stability and reliability of power supply and meeting the power demand of different user groups. The present invention effectively improves the regulation flexibility and adaptability of the distribution network in multiple scenarios through intelligent data acquisition and analysis, personalized power consumption feature identification, multi-objective strategy formulation and optimized decision-making selection, and uses simpler final regulation decisions to meet the power consumption needs of user groups in various scenarios in the distribution network as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of a forced power-off method based on a surgical robot according to an embodiment of the present invention;

[0063] Figure 2 This is a unit communication connection diagram of a main control system in another embodiment of the present invention. DETAILED DESCRIPTION

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0065] Combination Figure 1 As shown, the present invention provides an interactive control method for a multi-scenario distribution network, comprising:

[0066] Obtain real-time power distribution data of all scenario nodes in the distribution network.

[0067] Specifically, in the interactive control method of the multi-scenario distribution network, this is usually achieved through smart meters, sensors and data acquisition systems to ensure the accuracy and real-time nature of the data. It should be noted that the data collection frequency, data transmission method and data storage format need to be standardized. In an embodiment, smart meters deployed at key nodes of the distribution network are provided to obtain real-time distribution data of all scenario nodes in the distribution network.

[0068] The real-time power distribution data is analyzed to obtain the current power consumption data of each user group under all the scene nodes.

[0069] Specifically, in-depth analysis of real-time power distribution data is performed to determine the current power consumption of each user group. Generally, data preprocessing is required for real-time power distribution data, such as denoising and outlier removal, and then statistical analysis methods are applied to identify power consumption patterns. In a preferred embodiment of the present invention, a clustering algorithm can be used to classify user groups and analyze the power consumption characteristics of each group.

[0070] Predictions are made based on the current electricity usage data to obtain future electricity usage data of the user group within a preset future time period.

[0071] Specifically, based on the current electricity consumption data, the electricity consumption data of the user group in the future preset time period is predicted. This involves the selection of the prediction model, the setting of parameters and the accuracy evaluation of the prediction results. First, the prediction model is determined, such as ARIMA, neural network, etc., then the model is trained based on historical data, and finally the model is used to predict future electricity consumption data. It is worth mentioning that the accuracy of the prediction model needs to be evaluated regularly and adjusted as needed.

[0072] According to the future electricity consumption data in the preset future time period, the electricity consumption characteristics of each user group under all the scene nodes are obtained, and the electricity consumption characteristics include the electricity consumption pattern, peak and valley time periods and load trends of the user group.

[0073] Specifically, the load forecasting model is used to predict the power load in the future time period, taking into account seasonal changes and external factors such as weather conditions. In a preferred embodiment of the present invention, the power consumption pattern can be modeled by a machine learning algorithm to classify and identify the power consumption behavior of different user groups. Key power consumption features such as peak and valley periods, load fluctuations, and power consumption stability are extracted from the analysis results and their accuracy is ensured through cross-validation. Finally, these features are integrated and applied to formulate a control strategy that meets user needs.

[0074] According to the electricity consumption pattern, the peak and valley time periods and the load trend of each user group, the control target of each user group under all the scene nodes in the preset future time period is determined.

[0075] Specifically, it is necessary to first comprehensively analyze the key electricity consumption characteristics of the user group, such as the electricity consumption pattern, peak and valley time periods, and load trends. The electricity consumption pattern reflects the daily electricity consumption habits and behaviors of the user group, the peak and valley time periods reveal the specific time of peak and valley electricity consumption, and the load trend shows the trend of electricity demand over time. Based on these analysis results, the setting of control targets needs to consider multiple factors, including but not limited to the user's electricity demand, the operating status of the power grid, the energy supply situation, and the economic cost. The control objectives include improving energy utilization efficiency, ensuring power supply reliability, optimizing cost-effectiveness, balancing supply and demand, and promoting the use of renewable energy. Through this step, control targets that meet both their specific electricity consumption behaviors and the overall operation needs of the power grid can be formulated for the user group under each scenario node.

[0076] The control target of the user group and the current power consumption data of the user group are input into a multi-objective strategy model to obtain a sub-control decision for each of the user groups.

[0077] Specifically, control objectives such as minimizing costs, maximizing energy efficiency, and ensuring power supply stability are quantified into computable objective functions. Subsequently, these objectives are combined with the real-time electricity consumption data of the user group to form the input parameters of the multi-objective strategy model. In a preferred embodiment of the present invention, a genetic algorithm can be used to process these parameters while considering the trade-offs and conflicts between different objectives. In this way, the model is able to generate a series of feasible sub-control decisions for each user group, each of which aims to achieve a specific control goal while taking into account the overall operating efficiency of the power grid and the specific needs of the users.

[0078] According to the sub-regulation decision of each user group, the adaptability of each sub-regulation decision to the final regulation decision of the distribution network is determined.

[0079] Specifically, it is necessary to first define fitness indicators, which usually include grid stability, energy efficiency, cost-effectiveness, user satisfaction, and environmental impact. Then, these indicators are used to construct a fitness function, which quantitatively evaluates the contribution and impact of each sub-regulation decision on the overall goal of the distribution network. The evaluation process may involve simulating the implementation effect of sub-regulation decisions in the power grid, monitoring key parameters of power grid operation, such as voltage stability, frequency deviation, line load, etc., and collecting user feedback to measure user satisfaction. In this way, it is possible to identify which sub-regulation decisions most effectively balance the needs of different user groups and the long-term sustainable development of the power grid while meeting the requirements of power grid operation.

[0080] Multi-objective optimization is performed according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network.

[0081] Specifically, the fitness of each sub-regulation decision needs to be comprehensively evaluated first, and these evaluation results will serve as the input of the optimization algorithm. The multi-objective optimization process involves weighing the conflicts and priorities between different objectives, such as finding a balance between cost minimization and power supply reliability. Using methods such as Pareto optimization, genetic algorithms, or multi-criteria decision analysis, the algorithm will search for an optimal set of solutions that meet all or most of the objectives. These solutions will be ranked according to their fitness values ​​to determine which solutions best meet the overall needs and goals of the distribution network. The dynamic characteristics of the power grid and real-time feedback may also be considered in the optimization process to ensure the feasibility and effectiveness of the selected solution.

[0082] The power distribution network is controlled to distribute power according to the final regulation decision.

[0083] Specifically, the distribution network is controlled to distribute electricity according to the final control decision, including adjusting the operating parameters of the transformer, controlling the charging and discharging strategy of the energy storage system, optimizing the allocation of power generation resources, and implementing demand response measures. The automated control system will automatically adjust the grid equipment according to these instructions to ensure a balance between power supply and demand.

[0084] The interactive control method of the multi-scenario distribution network of the present invention accurately grasps the current power consumption of different user groups by acquiring and analyzing the power consumption data of all scenario nodes in the distribution network in real time. Then, the prediction technology is used to predict the power demand of the user group in the future time period to provide forward-looking information for the control decision. Not only the power demand is considered, but also the power consumption mode, peak and valley time period and load trend of the user are analyzed, so as to customize the personalized power consumption characteristics for each user group. The identification of these characteristics provides a more accurate basis for formulating the control target, so that the control strategy can better meet the specific needs of different user groups. Through the multi-objective strategy model, the control target of each user group is combined with the current power consumption data to generate a sub-control decision. This method not only considers a single goal, but also integrates multiple goals, such as cost, efficiency, reliability, etc., so as to achieve the optimal configuration of resources while meeting diversified needs. Through fitness evaluation and multi-objective optimization, the method can select the optimal final control decision from multiple possible sub-control decisions, ensuring that the distribution network can achieve cost-effectiveness while also ensuring the stability and reliability of power supply and meeting the power demand of different user groups. The present invention effectively improves the regulation flexibility and adaptability of the distribution network in multiple scenarios through intelligent data acquisition and analysis, personalized power consumption feature identification, multi-objective strategy formulation and optimized decision-making selection, and uses simpler final regulation decisions to meet the power consumption needs of user groups in various scenarios in the distribution network as much as possible.

[0085] In the embodiment of the present invention, the analysis is performed based on the real-time power distribution data to obtain the current power consumption data of each user group under all the scene nodes, including:

[0086] Acquire power consumption indicators of all the scene nodes according to the real-time power distribution data, wherein the power consumption indicators include current power, voltage, current and power of the scene nodes;

[0087] Determine the current power consumption data of the user group according to the power consumption index combined with the power proportion of each of the user groups under the scenario node;

[0088] wherein, according to the voltage and current of the scene node, the total power of the scene node is determined;

[0089] Determining the current power usage of the user group according to the total power and the power proportion;

[0090] The current power usage data of the user group is determined according to the current power usage.

[0091] In this embodiment, the power consumption indicators of all scene nodes, including power, voltage, current and power, are collected by smart meters and sensors. These indicators provide comprehensive operating status information for the distribution network. Using these indicators, the total power of each scene node can be calculated, that is, obtained by multiplying the voltage and current. Subsequently, according to the power share of each user group under the scene node, the total power is proportionally distributed to each user group, thereby determining the current power usage of each user group. Specifically, the total power of the scene node is obtained by multiplying the voltage and current, and the current power usage of the user group is the product of the total power and its power share. Through this step, the power consumption of each user group can be accurately quantified, providing accurate data support for subsequent regulation decisions.

[0092] The interactive control method of the multi-scenario distribution network of the present invention provides real-time and accurate electricity consumption information by acquiring the current electricity consumption data of each user group under all scenario nodes, so that the power grid operator can quickly respond to changes in the operating status of the power grid and adjust the power supply strategy in time to meet the needs of different user groups.

[0093] In the embodiment of the present invention, the step of performing prediction based on the current power consumption data to obtain future power consumption data of the user group within a preset future time period includes:

[0094] Acquiring influencing factor data within the preset future time period, the influencing factor data including the geographical location and climate data of the user group;

[0095] Determine a corresponding prediction model according to the current power consumption data, the geographical location and the climate data, and input the current power consumption data, the geographical location and the climate data into the prediction model;

[0096] Batch prediction is performed on each of the user groups through the prediction model to obtain the future electricity consumption data of the user group in the preset future time period.

[0097] In this embodiment, firstly, data of key factors that may affect the electricity consumption pattern in a preset future time period are collected, such as the user's geographical location and climatic conditions. These data are essential for understanding the seasonal and regional differences in electricity consumption behavior. Then, a suitable prediction model, such as time series analysis, machine learning model or deep learning model, is selected or constructed in combination with the current electricity consumption data, geographical location and climate data. After these data are input into the model, batch predictions can be made for each user group to obtain the user electricity consumption prediction data in the preset future time period. For example, if the geographical location of an industrial area is located in an area with a large population, and the residential population and the industrial area are set as the same scene, but its electricity consumption pattern may conflict with the electricity consumption of the residential population, because more electricity is needed to maintain industrial operation, but at the same time, the electricity consumption of the residential population does not require such a large distribution of electricity. The prediction model will consider this factor to adjust its prediction.

[0098] The interactive control method of the multi-scenario distribution network of the present invention predicts future electricity consumption data for a preset future time period, so that the grid operator can foresee and prepare for fluctuations in electricity demand in advance. Secondly, by predicting electricity consumption data under different geographical locations and climatic conditions, the grid operator can more accurately plan grid resource allocation, optimize power generation and transmission plans, and reduce the risk of energy waste or power shortage due to inaccurate demand forecasts.

[0099] In the embodiment of the present invention, obtaining the power consumption characteristics of each user group under all the scene nodes according to the future power consumption data in the preset future time period includes:

[0100] The user group is grouped according to the future electricity consumption data of the user group in the preset future time period through the objective function of the K-means clustering algorithm, and the user group is divided into a plurality of clusters;

[0101] Place Description The objective function is:

[0102] J= ;

[0103] in, is the number of clusters, is the user group of the i-th cluster, is the center point of the power consumption feature of the i-th cluster, x is the future power consumption data of the user group, and J is the sum of the squares of the distances from all the clusters to the center point of the power consumption feature;

[0104] Feature extraction is performed based on the future electricity usage data of the user group of each cluster to obtain the electricity usage features of the user group.

[0105] In this embodiment, the K-means clustering algorithm is first used to group user groups to identify user groups with similar electricity consumption patterns. The K-means algorithm achieves clustering by minimizing the objective function J, which calculates the sum of the squares of the distances from all user groups to their cluster centers. In this process, represents the number of clusters, represents the set of user groups in the i-th cluster, is the center point of the electricity consumption characteristics of the ith cluster, and x is the future electricity consumption data of the user group. For example, if the user group in a residential area is divided into two clusters based on their electricity consumption data, one cluster may represent the group with peak electricity consumption at night, and the other cluster may represent the group with more electricity consumption during the day. In this way, user groups with different electricity consumption behavior characteristics can be identified. For clusters with peak electricity consumption at night, a demand response strategy is implemented to encourage users to use more electricity during non-peak hours, while for clusters with more electricity consumption during the day, the power supply capacity during the day can be increased.

[0106] The interactive control method of the multi-scenario distribution network of the present invention can identify user groups with similar electricity consumption patterns, so that grid operators can more accurately predict and plan the allocation of power resources, thereby improving the operation efficiency of the grid. Secondly, the clustering results can help operators design customized services and control strategies to meet the specific needs of different user groups.

[0107] In the embodiment of the present invention, the feature extraction is performed based on the future power consumption data of each clustered user group to obtain the power consumption features of the user group, and the power consumption features include the power consumption mode, peak and valley time periods, and load trends of the user group, including:

[0108] Obtaining a load curve of the user group in the preset future time period according to the future power consumption data of the clustered user group;

[0109] The electricity consumption behavior of the user group in the preset future time period is predicted according to the load curve to obtain the electricity consumption pattern, the peak and valley time periods and the load trend of the user group in the preset future time period.

[0110] In this embodiment, the load curve of the user group in the future preset time period is first analyzed, and the load curve is analyzed by aggregating the power consumption data of each user group. Among them, the load curve shows the power consumption of the user group in different time periods, so that the power consumption pattern, peak and valley time periods and load trends can be identified. For example, if an industrial user group shows a higher power load during the day on weekdays, and the load decreases at night and on weekends, then it can be predicted that the user group has a clear power consumption pattern with peaks on weekdays and valleys on weekends. By analyzing the load curve, the power consumption behavior of the user group in a specific time period can be further predicted, including peak and valley time periods and load change trends.

[0111] The interactive control method of the multi-scenario distribution network of the present invention enables the grid operator to plan power resources in advance and optimize power generation and transmission plans to meet the power demand of different user groups by accurately predicting the load curve and power consumption pattern.

[0112] In the embodiment of the present invention, determining the control target of each user group in the preset future time period under all the scenario nodes according to the power consumption pattern, the peak and valley time periods, and the load trend of each user group includes:

[0113] According to the power demand of the user group in the preset future time period, obtaining the expected power distribution data in the preset future time period;

[0114] The control target is obtained according to the power consumption pattern of the user group, the peak and valley time periods, and the load trend and the expected power distribution data corresponding to each of the user groups.

[0115] In this embodiment, it is first necessary to understand the power demand of the user group during the time period, which is usually obtained by analyzing its power consumption pattern, peak and valley time periods and load trends. Then, based on these power consumption characteristics, the expected power distribution data can be derived, that is, the expected power supply of the user group in different time periods. For example, if the user group in a residential area shows high power demand between 7 pm and 10 pm, this may indicate that this period is a peak power consumption period. The grid operator combines this information with the power consumption pattern of the residential area and sets the expected power distribution data to ensure that there is enough power supply to meet the demand during peak hours. Then, by comparing the actual power distribution data with the expected power distribution data, the regulation target can be determined, which may include increasing power generation during peak hours or adjusting the grid operation parameters to meet high demand. For example, if the user group in a commercial area shows higher power demand on weekends, the grid operator can arrange more power supply during these periods to avoid potential power shortages.

[0116] The interactive control method of the multi-scenario distribution network of the present invention determines the control target of each user group under all scenario nodes in a preset future time period. The distribution network can respond to the continuous changes in electricity demand more flexibly while maintaining efficient and economical operation, thereby improving the stability and reliability of the power grid.

[0117] In the embodiment of the present invention, the step of inputting the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model to obtain a sub-control decision for each user group includes:

[0118] Inputting the regulation target of the user group and the current power consumption data of the user group into a multi-objective strategy model;

[0119] The current power consumption data and the control target are integrated to obtain an integrated feature vector;

[0120] Assigning a weight to each feature vector in the integrated feature vector, and then solving the multi-objective strategy model through an optimization algorithm to obtain a Pareto optimal solution set;

[0121] Screening is performed according to the Pareto optimal solution set to obtain the sub-regulation decision of the user group.

[0122] In this embodiment, the control objectives of the user group are combined with the current power consumption data and input into the multi-objective strategy model. This step involves fusing the specific requirements of the control objectives, such as minimizing costs, maximizing energy efficiency, ensuring power supply reliability, etc., with real-time power consumption data, such as current power, voltage, current and power, to form a comprehensive feature vector. For example, if a user group in an industrial area needs stable power supply during peak hours in the morning and evening, its control objectives may include maintaining voltage stability and reducing the number of power outages during these periods. Combining these goals with the current power consumption data, the feature vector formed will reflect these requirements. Next, weights are assigned to each feature in the feature vector, reflecting the priority and importance of different goals, and the multi-objective strategy model is solved by an optimization algorithm, such as a genetic algorithm or a multi-criteria decision analysis, to obtain a set of Pareto optimal solutions. The Pareto optimal solution set represents a series of solutions that balance multiple goals, and each solution is non-dominated, that is, a certain goal cannot be further optimized without sacrificing other goals. For example, if a user group in a commercial area has a surge in electricity demand on weekends, the Pareto optimal solution set may include adjusting the charging and discharging strategy of energy storage equipment to meet the peak demand without increasing costs. By further screening the Pareto optimal solution set, sub-regulation decisions for each user group can be obtained based on the actual grid operation status and the specific details of user needs.

[0123] The interactive control method of the multi-scenario distribution network of the present invention finds the optimal solution by balancing multiple control objectives to meet the diverse needs of different user groups. Secondly, by assigning weights to each feature, the grid operator can adjust the priority of the target according to the actual situation and implement a more flexible control strategy.

[0124] In the embodiment of the present invention, determining the adaptability of each sub-regulation decision to the final regulation decision of the distribution network according to the sub-regulation decision of each user group includes:

[0125] Encoding the sub-regulation decision to obtain a binary code of the sub-regulation decision;

[0126] The adaptability of the sub-regulation decision to the final regulation decision of the power distribution network is determined by a fitness function according to the binary encoding of the sub-regulation decision.

[0127] In this embodiment, each sub-regulation decision is first encoded, generally in the form of binary encoding. This encoding method converts the parameters and variables of the sub-regulation decision into a format that can be processed by a computer, which is convenient for subsequent fitness evaluation. For example, if the sub-regulation decision of a residential area includes adjusting the voltage level and changing the load distribution, these decisions can be encoded as a binary string, and each bit represents a specific regulation parameter. Secondly, the fitness function is used to evaluate these binary-coded sub-regulation decisions. The fitness function is an evaluation indicator that quantifies the pros and cons of each sub-regulation decision based on the operating objectives and constraints of the distribution network. The fitness function may consider multiple aspects such as grid stability, power supply reliability, economic cost and user satisfaction. For example, if the sub-regulation decision of an industrial area can reduce costs while ensuring power supply reliability, then this decision will score higher in the fitness function. Through the binary encoding of sub-regulation decisions and the application of fitness functions, the grid operator can quantify the contribution of each decision to the final regulation decision of the distribution network, and select and optimize accordingly.

[0128] The interactive control method of the multi-scenario distribution network of the present invention allows the grid operator to dynamically adjust the priority and weight of the sub-control decision according to the actual operation status and objectives of the grid through the use of the fitness function, so as to achieve more flexible and personalized control. At the same time, it helps to identify and optimize the key control decisions that have a significant impact on the overall performance of the grid, thereby improving the operation efficiency and reliability of the grid.

[0129] In an embodiment of the present invention, performing multi-objective optimization according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network includes:

[0130] Initializing according to the sub-regulation decision of the user group to obtain an initial population;

[0131] Selecting the sub-regulation decision from the initial population according to the fitness, and iterating the selected sub-regulation decision through a crossover mutation operation to form a new population;

[0132] Updating the fitness of the sub-regulation decision in the new population to obtain the updated fitness;

[0133] When the updated fitness reaches a preset threshold, the Pareto optimal solution set of the final control decision is obtained from the new population formed in the last iteration;

[0134] According to the Pareto optimal solution set, the final control decision is obtained.

[0135] In this embodiment, the initial population is first obtained based on the sub-regulation decision initialization of the user group, which is regarded as a set of possible regulation strategies. Then, the sub-regulation decisions with better performance are selected from the initial population through fitness evaluation, and these strategies are iteratively optimized using the crossover mutation operation in the genetic algorithm to form a new population. For example, if the sub-regulation decision of a commercial district includes adjusting the charging and discharging strategies of loads and energy storage devices, these strategies will generate new regulation combinations through crossover mutation to explore better solutions. During the iteration process, the fitness of the sub-regulation decisions in the new population is updated to reflect their performance in meeting the grid regulation objectives. Fitness evaluation may consider multiple objectives such as cost, efficiency, and reliability. As the iteration proceeds, the fitness evaluation helps identify regulation strategies that are getting closer and closer to the optimal solution. When the fitness reaches a preset threshold, it indicates that a satisfactory optimization level has been approached or reached. At this time, the Pareto optimal solution set is extracted from the new population formed by the last iteration. The Pareto optimal solution set represents the optimal regulation strategy set that strikes a balance between multiple objectives. For example, if an industrial zone's control strategy finds the best balance between reducing costs and improving energy efficiency, this strategy will be included in the Pareto optimal solution set. Finally, the final control decision is selected from the Pareto optimal solution set, which takes all objectives into consideration to achieve the overall optimization of grid operation.

[0136] The interactive control method of the multi-scenario distribution network of the present invention enables the grid operator to continuously improve the control strategy through iterative optimization to adapt to the changes in grid operating conditions. Secondly, the use of fitness evaluation and Pareto optimal solution set ensures that the control decision achieves the best balance between multiple objectives, thereby improving the operating efficiency and economy of the grid.

[0137] Combination Figure 2As shown, the present invention also provides an interactive control system for a multi-scenario distribution network, comprising:

[0138] A data acquisition unit, used to acquire real-time power distribution data of all scene nodes in the distribution network;

[0139] A first analysis unit, configured to analyze the real-time power distribution data to obtain current power consumption data of each user group under all the scenario nodes;

[0140] A prediction unit, configured to make a prediction based on the current power consumption data to obtain future power consumption data of the user group within a preset future time period;

[0141] A second analysis unit is used to obtain the power consumption characteristics of each user group under all the scene nodes according to the future power consumption data in the preset future time period, wherein the power consumption characteristics include the power consumption mode, peak and valley time periods, and load trends of the user group;

[0142] Determine the control target of each user group under all the scenario nodes in the preset future time period according to the power consumption pattern, the peak and valley time periods and the load trend of each user group;

[0143] A first decision unit, configured to input the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model to obtain a sub-control decision for each of the user groups;

[0144] Determining, according to the sub-regulation decision of each user group, the adaptability of each sub-regulation decision to the final regulation decision of the distribution network;

[0145] A second decision-making unit, configured to perform multi-objective optimization according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network;

[0146] A control unit is used to control the power distribution network to distribute power according to the final control decision.

[0147] The interactive control system of the multi-scenario distribution network of the present invention accurately grasps the current power consumption of different user groups by acquiring and analyzing the power consumption data of all scene nodes in the distribution network in real time. Then, the prediction technology is used to predict the power demand of the user group in the future time period to provide forward-looking information for the control decision. Not only the power demand is considered, but also the power consumption mode, peak and valley time period and load trend of the user are analyzed, so as to customize the personalized power consumption characteristics for each user group. The identification of these characteristics provides a more accurate basis for formulating the control target, so that the control strategy can better meet the specific needs of different user groups. Through the multi-objective strategy model, the control target of each user group is combined with the current power consumption data to generate a sub-control decision. This method not only considers a single goal, but also integrates multiple goals, such as cost, efficiency, reliability, etc., so as to achieve the optimal configuration of resources while meeting diversified needs. Through fitness evaluation and multi-objective optimization, the method can select the optimal final control decision from multiple possible sub-control decisions, ensuring that the distribution network can achieve cost-effectiveness while also ensuring the stability and reliability of power supply and meeting the power demand of different user groups. The present invention effectively improves the regulation flexibility and adaptability of the distribution network in multiple scenarios through intelligent data acquisition and analysis, personalized power consumption feature identification, multi-objective strategy formulation and optimized decision-making selection, and uses simpler final regulation decisions to meet the power consumption needs of user groups in various scenarios in the distribution network as much as possible.

[0148] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0149] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0150] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. An interactive control method for a multi-scenario distribution network, characterized in that: include: Obtain real-time power distribution data of all scene nodes in the distribution network; Analyze the real-time power distribution data to obtain current power consumption data of each user group under all the scenario nodes; Perform prediction based on the current electricity consumption data to obtain future electricity consumption data of the user group within a preset future time period; According to the future electricity consumption data in the preset future time period, the electricity consumption characteristics of each user group under all the scene nodes are obtained, and the electricity consumption characteristics include the electricity consumption mode, peak and valley time periods and load trends of the user group; Determine the control target of each user group under all the scenario nodes in the preset future time period according to the power consumption pattern, the peak and valley time periods and the load trend of each user group; Inputting the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model to obtain a sub-control decision for each user group; specifically comprising: inputting the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model; The current power consumption data and the control target are integrated to obtain an integrated feature vector; Assigning a weight to each feature vector in the integrated feature vector, and then solving the multi-objective strategy model through an optimization algorithm to obtain a Pareto optimal solution set; Screening is performed according to the Pareto optimal solution set to obtain the sub-regulation decision of the user group; According to the sub-regulation decision of each user group, determining the adaptability of each sub-regulation decision to the final regulation decision of the distribution network; specifically comprising: encoding the sub-regulation decision to obtain a binary code of the sub-regulation decision; Determining the fitness of the sub-regulation decision to the final regulation decision of the distribution network according to the binary code of the sub-regulation decision by means of a fitness function; Performing multi-objective optimization according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network; specifically comprising: initializing according to the sub-regulation decisions of the user group to obtain an initial population; Selecting the sub-regulation decision from the initial population according to the fitness, and iterating the selected sub-regulation decision through a crossover mutation operation to form a new population; Updating the fitness of the sub-regulation decision in the new population to obtain the updated fitness; When the updated fitness reaches a preset threshold, the Pareto optimal solution set of the final control decision is obtained according to the new population formed in the last iteration; According to the Pareto optimal solution set, the final control decision is obtained; The power distribution network is controlled to distribute power according to the final regulation decision.

2. The interactive control method of multi-scenario distribution network according to claim 1 is characterized in that: The analyzing according to the real-time power distribution data to obtain the current power consumption data of each user group under all the scene nodes includes: Acquire power consumption indicators of all the scene nodes according to the real-time power distribution data, wherein the power consumption indicators include current power, voltage, current and power of the scene nodes; Determine the current power consumption data of the user group according to the power consumption index combined with the power proportion of each of the user groups under the scenario node; wherein, according to the voltage and current of the scene node, the total power of the scene node is determined; Determining the current power usage of the user group according to the total power and the power proportion; The current power usage data of the user group is determined according to the current power usage.

3. The interactive control method of multi-scenario distribution network according to claim 2 is characterized in that: The predicting according to the current power consumption data to obtain future power consumption data of the user group within a preset future time period includes: Acquiring influencing factor data within the preset future time period, the influencing factor data including the geographical location and climate data of the user group; Determine a corresponding prediction model according to the current power consumption data, the geographical location and the climate data, and input the current power consumption data, the geographical location and the climate data into the prediction model; Batch prediction is performed on each of the user groups through the prediction model to obtain the future electricity consumption data of the user group in the preset future time period.

4. The interactive control method of multi-scenario distribution network according to claim 2 is characterized in that: The obtaining, according to the future electricity consumption data in the preset future time period, electricity consumption characteristics of each user group under all the scene nodes includes: The user group is grouped according to the future electricity consumption data of the user group in the preset future time period through the objective function of the K-means clustering algorithm, and the user group is divided into a plurality of clusters; Place Description The objective function is: J＝ ； in, is the number of clusters, is the user group of the i-th cluster, is the center point of the power consumption feature of the i-th cluster, x is the future power consumption data of the user group, and J is the sum of the squares of the distances from all the clusters to the center point of the power consumption feature; Feature extraction is performed based on the future electricity usage data of the user group of each cluster to obtain the electricity usage features of the user group.

5. The interactive control method of multi-scenario distribution network according to claim 4 is characterized in that: The feature extraction is performed based on the future power consumption data of each clustered user group to obtain the power consumption features of the user group, wherein the power consumption features include the power consumption mode, peak and valley time periods, and load trends of the user group, including: Obtaining a load curve of the user group in the preset future time period according to the future power consumption data of the clustered user group; The electricity consumption behavior of the user group in the preset future time period is predicted according to the load curve to obtain the electricity consumption pattern, the peak and valley time periods and the load trend of the user group in the preset future time period.

6. The interactive control method of multi-scenario distribution network according to claim 1 is characterized in that: Determining the control target of each user group in the preset future time period under all the scene nodes according to the power consumption pattern, the peak and valley time periods, and the load trend of each user group includes: According to the power demand of the user group in the preset future time period, obtaining the expected power distribution data in the preset future time period; The control target is obtained according to the power consumption pattern of the user group, the peak and valley time periods, and the load trend and the expected power distribution data corresponding to each of the user groups.

7. An interactive control system for a multi-scenario distribution network, characterized in that: The interactive control method for a multi-scenario distribution network according to any one of claims 1 to 6, wherein the interactive control system for the multi-scenario distribution network comprises: A data acquisition unit, used to acquire real-time power distribution data of all scene nodes in the distribution network; A first analysis unit, configured to analyze the real-time power distribution data to obtain current power consumption data of each user group under all the scenario nodes; A prediction unit, configured to make a prediction based on the current power consumption data to obtain future power consumption data of the user group within a preset future time period; A second analysis unit is used to obtain the power consumption characteristics of each user group under all the scene nodes according to the future power consumption data in the preset future time period, wherein the power consumption characteristics include the power consumption mode, peak and valley time periods, and load trends of the user group; Determine the control target of each user group under all the scenario nodes in the preset future time period according to the power consumption pattern, the peak and valley time periods and the load trend of each user group; A first decision unit is used to input the control target of the user group and the current power consumption data of the user group into a multi-objective strategy model to obtain a sub-control decision for each user group; specifically comprising: inputting the control target of the user group and the current power consumption data of the user group into the multi-objective strategy model; The current power consumption data and the control target are integrated to obtain an integrated feature vector; Assigning a weight to each feature vector in the integrated feature vector, and then solving the multi-objective strategy model through an optimization algorithm to obtain a Pareto optimal solution set; Screening is performed according to the Pareto optimal solution set to obtain the sub-regulation decision of the user group; According to the sub-regulation decision of each user group, determining the adaptability of each sub-regulation decision to the final regulation decision of the distribution network; specifically comprising: encoding the sub-regulation decision to obtain a binary code of the sub-regulation decision; Determining the fitness of the sub-regulation decision to the final regulation decision of the distribution network according to the binary code of the sub-regulation decision by means of a fitness function; The second decision unit is used to perform multi-objective optimization according to the fitness of each of the sub-regulation decisions to obtain the final regulation decision of the distribution network; specifically comprising: initializing according to the sub-regulation decisions of the user group to obtain an initial population; Selecting the sub-regulation decision from the initial population according to the fitness, and iterating the selected sub-regulation decision through a crossover mutation operation to form a new population; Updating the fitness of the sub-regulation decision in the new population to obtain the updated fitness; When the updated fitness reaches a preset threshold, the Pareto optimal solution set of the final control decision is obtained according to the new population formed in the last iteration; According to the Pareto optimal solution set, the final control decision is obtained; A control unit is used to control the power distribution network to distribute power according to the final control decision.

Images