Aggregator level dynamic aggregation and control method suitable for industrial and commercial loads

Abstract

The embodiment of the invention discloses an aggregator level dynamic aggregation and control method suitable for industrial and commercial loads. The method comprises the following steps: acquiring historical power consumption data of a plurality of industrial and commercial users; extracting load characteristics of the historical power consumption data, and performing data aggregation processing according to the plurality of load characteristics to obtain a plurality of target aggregation sets; calculating the similarity between the target cluster set and a preset load type; if the similarity is greater than a first threshold value, scheduling distributed power resources to a plurality of industrial and commercial users corresponding to the target clustering set by adopting a regulation and control strategy corresponding to the preset load type; through fine analysis and dynamic aggregation of industrial and commercial user load characteristics, accurate matching of regulation and control strategies and efficient scheduling of electric power resources are realized, the rationality of load regulation and control and the utilization efficiency of the electric power resources can be improved, and safe and stable operation of a power grid is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of power distribution system regulation and control, and in particular to a method for dynamic aggregation and control of aggregators at the aggregator level adapted to industrial and commercial loads. Background Art

[0002] With the development of the Energy Internet and the ongoing reform of the power market, the electricity demand and consumption behavior of industrial and commercial users have become increasingly complex and diverse, posing new challenges to the dispatch and management of power systems. Industrial and commercial loads are typically characterized by significant volatility and uncertainty, especially during peak demand periods and seasonal transitions, when load fluctuations are more pronounced. This load uncertainty puts pressure on the stable operation of the power grid and increases the difficulty of power dispatch and resource allocation.

[0003] Currently, when responding to the diverse load demands and changing load characteristics of industrial and commercial users, response speed and dispatch efficiency are often low. It is difficult to accurately identify and distinguish different types of load patterns, resulting in a deviation between dispatch strategies and actual demand, which affects the optimal allocation of power resources. Therefore, how to improve the rationality of load regulation and the efficiency of power resource utilization has become a pressing technical issue. Summary of the Invention

[0004] The main purpose of this application is to provide an aggregator-level dynamic aggregation and control method that is suitable for industrial and commercial loads, so as to improve the rationality of load regulation and the utilization efficiency of power resources.

[0005] To achieve the above objectives, the present application provides, in a first aspect, a method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads, the method comprising:

[0006] Obtain historical electricity consumption data for multiple industrial and commercial users;

[0007] Extracting load characteristics of the historical electricity consumption data, performing data aggregation processing based on a plurality of the load characteristics, and obtaining a plurality of target aggregation sets;

[0008] Calculating the similarity between the target cluster set and the preset load type;

[0009] If the similarity is greater than a first threshold, the control strategy corresponding to the preset load type is adopted to dispatch distributed power resources to multiple industrial and commercial users corresponding to the target cluster set.

[0010] A second aspect of the present application provides an aggregator-level dynamic aggregation and control device adapted to industrial and commercial loads, comprising:

[0011] A data acquisition module is used to obtain historical electricity consumption data of multiple industrial and commercial users;

[0012] A characteristic extraction module, configured to extract load characteristics of the historical power consumption data;

[0013] an aggregation processing module, configured to perform data aggregation processing according to the plurality of load characteristics to obtain a plurality of target aggregation sets;

[0014] A similarity calculation module, configured to calculate the similarity between the target cluster set and a preset load type;

[0015] The control strategy execution module is used to adopt the control strategy corresponding to the preset load type to dispatch distributed power resources to multiple industrial and commercial users corresponding to the target cluster set if the similarity is greater than a first threshold.

[0016] A third aspect of the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the first aspect and any possible implementation thereof.

[0017] A fourth aspect of the present application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the method described in the first aspect.

[0018] The present application provides an aggregator-level dynamic aggregation and control method that is suitable for industrial and commercial loads. The method obtains historical electricity consumption data of multiple industrial and commercial users, extracts the load characteristics of the historical electricity consumption data, performs data aggregation processing based on the multiple load characteristics, and obtains multiple target aggregation sets; calculates the similarity between the target cluster set and a preset load type; if the similarity is greater than a first threshold, adopts the control strategy corresponding to the preset load type to dispatch distributed power resources to the multiple industrial and commercial users corresponding to the target cluster set; through the refined analysis and dynamic aggregation of the load characteristics of industrial and commercial users, accurate matching of the control strategy and efficient scheduling of power resources are achieved, which can improve the rationality of load control and the utilization efficiency of power resources, and ensure the safe and stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] in:

[0021] Figure 1A flowchart of a method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads provided in an embodiment of the present application;

[0022] Figure 2 A schematic diagram of a typical daily load and distributed photovoltaic energy curve for a certain park provided in an embodiment of the present application;

[0023] Figure 3 A schematic diagram of control results at different adjustable load ratios provided in an embodiment of the present application;

[0024] Figure 4 A schematic diagram of another control result with different adjustable load ratios provided in an embodiment of the present application;

[0025] Figure 5 A schematic diagram of the structure of an aggregator-level dynamic aggregation and control device adapted to industrial and commercial loads provided in an embodiment of the present application;

[0026] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0028] The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0029] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0030] Demand side management (DSM) involved in the embodiments of this application is an important part of power system optimization and plays a key role in effectively utilizing renewable energy, reducing power system operating costs, and improving power supply reliability. In this context, aggregators, as an important role in demand side management, can aggregate decentralized load resources into a controllable virtual power plant (VPP), provide auxiliary services to the power grid, and obtain additional benefits by participating in power market transactions.

[0031] The following describes the related concepts or functional implementations that may be involved in the methods in the embodiments of the present application:

[0032] Distributed control systems: Distributed control technology can distribute load management tasks to control centers in different regions, reducing the risk of single points of failure. Edge computing technology can also be used to offload some computing tasks to edge nodes close to the load points, reducing data transmission delays and improving control response speed.

[0033] Multi-objective optimization: Load aggregation and scheduling require a trade-off between multiple objectives, including cost-effectiveness, stability, and user satisfaction. Multi-objective optimization techniques, such as genetic algorithms and particle swarm optimization, can help aggregators find the optimal balance between these objectives and maximize the overall benefits of load scheduling.

[0034] Demand response (DR) is an important strategy for dynamic load control. Aggregators can flexibly adjust loads through cooperation with industrial and commercial users to achieve grid balance and economic benefits.

[0035] Market-based demand response: Aggregators monitor electricity market price fluctuations in real time and send regulatory signals to industrial and commercial users. Users can then adjust their electricity usage based on this price information, reducing load when prices are high and increasing load when prices are low, thereby participating in demand response.

[0036] Incentive mechanism: To encourage industrial and commercial users to participate in load dispatch, aggregators can provide economic incentives, such as giving users electricity price discounts or subsidies during the demand response period, helping users reduce electricity costs while alleviating peak pressure on the power grid.

[0037] A virtual power plant (VPP) is a virtual resource that integrates distributed energy and dispatchable loads. Through dynamic aggregation control technology, a virtual power plant can effectively integrate various types of load resources.

[0038] Load aggregation and distributed energy integration: Aggregators can integrate distributed photovoltaic, wind power, and energy storage equipment with industrial and commercial loads to form virtual power plants. This approach not only smooths load fluctuations but also maximizes the use of renewable energy and reduces dependence on traditional power plants.

[0039] Virtual power plants participate in electricity markets: Virtual power plants can participate in electricity market transactions as a whole, providing ancillary services and balancing grid loads. By rationally dispatching loads and distributed resources, virtual power plants can generate more revenue in the electricity market and improve economic efficiency.

[0040] Energy storage: By integrating energy storage, aggregators can store excess electricity when electricity prices are low and release it during peak hours, reducing peak loads. This approach not only helps commercial and industrial users save on electricity costs but also reduces pressure on the grid during peak hours.

[0041] Combining demand response with energy storage: Energy storage systems combined with demand response strategies can further enhance load regulation flexibility. For example, aggregators can precisely regulate the load of industrial and commercial users by controlling the charging and discharging behavior of energy storage devices.

[0042] In addition, the method or system in the embodiments of the present application can be applied in combination with blockchain technology, edge computing, and Internet of Things technology.

[0043] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0044] See also Figure 1 , which is a flow chart of a method for dynamic aggregation and control of aggregators at the aggregator level adapted to industrial and commercial loads provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0045] 101. Obtain historical electricity consumption data for multiple industrial and commercial users.

[0046] The execution subject of the method in the embodiment of the present application can be an aggregator-level dynamic aggregation and control device that adapts to industrial and commercial loads. In actual applications, it can be implemented on an electronic device. The above-mentioned electronic device can be a terminal device such as a computer, or it can be implemented in a system, server, etc.

[0047] The collected historical electricity consumption data can also be pre-processed as needed, which is not limited in this embodiment of the present application.

[0048] 102. Extract the load characteristics of the historical electricity consumption data, perform data aggregation processing based on multiple load characteristics, and obtain multiple target aggregation sets.

[0049] The above-mentioned load characteristics are used to represent the distribution characteristics of historical electricity consumption data, including but not limited to one indicator or a combination of multiple indicators such as volatility index, daily variation characteristic index, periodicity index, seasonality index and randomness index.

[0050] The method for extracting the above load characteristics will be described in detail later. The following first introduces the data aggregation process in this application.

[0051] Users with the same or similar power load characteristics have the same or similar power demands. Therefore, in order to better dispatch power resources, this application aggregates multiple load characteristics to achieve aggregate power dispatch in the future.

[0052] In order to better dispatch power resources to users with the same or similar power load characteristics, it is necessary to perform data aggregation based on the load characteristics of different users to obtain multiple target aggregation sets. The specific logic is as follows:

[0053] 21. Construct multiple load characteristics into a load characteristic vector;

[0054] Specifically, multiple load characteristics may be coded to obtain multiple coded values, which are then constructed into a load characteristic vector in a preset order.

[0055] 22. Randomly extract a preset number of load characteristic vectors, and calculate the distances between the preset number of load characteristic vectors;

[0056] During the initial aggregation phase, a preset number of load feature vectors are randomly extracted. This number can be adjusted based on the dataset size and feature space dimensionality. After extraction, the distances between these load feature vectors are calculated using an appropriate distance metric (such as Euclidean distance or Manhattan distance). This distance measures the similarity between load feature vectors, with shorter distances indicating more similar characteristics.

[0057] 23. The load characteristic vectors whose distance is less than the second threshold are taken as the same initial aggregation set, and the load characteristic vectors whose distance is not less than the second threshold are taken as two initial aggregation sets;

[0058] The second threshold can be set as needed. When the distance between two sets of load characteristic vectors is less than the second threshold, it indicates that they have similar characteristics, and therefore they are grouped into the same initial cluster. If the distance is not less than the threshold, they are assigned to different initial clusters. This step lays the initial foundation for the subsequent clustering process.

[0059] 24. Calculate the distances between other load characteristic vectors and multiple initial aggregation sets in sequence;

[0060] Other load characteristic vectors refer to load characteristic vectors other than the above-mentioned preset number of load characteristic vectors.

[0061] 25. If the distance is less than a second threshold, then other load characteristic vectors with distances less than the threshold are included in the initial aggregation set;

[0062] 26. If the distance is not less than the second threshold, other load characteristic vectors with distances less than the threshold are used as new initial aggregation sets;

[0063] The remaining load characteristic vectors are then individually calculated relative to the defined initial clusters. If a vector's distance from an initial cluster is less than a second threshold, it is assigned to the corresponding cluster. If not, a new cluster is created based on the vector. This iterative process dynamically expands existing clusters or creates new ones, ensuring the proper classification of load characteristics.

[0064] 27. The multiple initial aggregation sets finally obtained are used as the multiple target aggregation sets.

[0065] After all load characteristic vectors have been calculated and classified as described above, the system ultimately generates multiple clusters. Each cluster represents a group of vectors with similar load characteristics, and these clusters are referred to as target clusters. In this way, load data can be effectively classified.

[0066] The above-described solution achieves efficient clustering of complex load characteristic data by constructing multiple load characteristic vectors into load characteristic vectors. First, a preset number of load characteristic vectors are randomly extracted and initially clustered based on the distance between them. When the distance is less than a second threshold, these load characteristic vectors are grouped into the same initial cluster set, ensuring that similar characteristics are clustered. For load characteristic vectors with a distance greater than or equal to the threshold, these vectors are assigned to different initial cluster sets, ensuring that loads with significantly different characteristics can be effectively distinguished. Based on this, the distances between other load characteristic vectors and the generated initial cluster sets are further calculated. Load characteristic vectors that meet the clustering criteria are dynamically incorporated into the corresponding initial cluster set, improving clustering accuracy and adaptability. Meanwhile, load characteristic vectors that do not meet the clustering criteria are reconstructed into new initial cluster sets, ensuring that each load characteristic is effectively clustered. The resulting multiple target cluster sets fully reflect the similarities and differences between different load characteristics, thereby improving data processing efficiency and accuracy, and are suitable for intelligent analysis of large-scale, complex load characteristic data.

[0067] In an optional implementation, different load characteristics have different calculation logics, and the specific logic is as follows:

[0068] 31. Calculate the first standard deviation of the plurality of historical electricity consumption data corresponding to each industrial and commercial user as a volatility indicator;

[0069] For each commercial and industrial user, the standard deviation of their historical electricity usage data is used to measure data volatility. Standard deviation is a key statistic that characterizes the degree of data dispersion and reflects fluctuations in electricity consumption over a given period. By calculating the first standard deviation of multiple historical electricity usage data points, we obtain a user's electricity usage volatility index, which can be used to identify the stability of a user's daily electricity consumption. For example, a user with high volatility may have electricity demand that fluctuates frequently with business activities.

[0070] 32. For each industrial and commercial user, extract a first median of the plurality of historical electricity usage data corresponding to the plurality of historical electricity usage data;

[0071] The median is a robust statistic that effectively reduces the impact of extreme values on analytical results. Therefore, extracting the median of historical electricity usage data can serve as a benchmark to help analyze electricity usage patterns. Compared to the mean, the median is less sensitive to abnormal fluctuations in data and is therefore a suitable benchmark for total electricity consumption, useful for assessing users' daily electricity usage habits.

[0072] 33. Divide the daily historical electricity consumption data into preset time periods, divide the total amount of first historical electricity consumption data corresponding to each of the multiple preset time periods by the first median, and obtain a daily variation characteristic index consisting of multiple values;

[0073] Daily historical electricity consumption data is divided into preset time periods, and the electricity consumption within each period is divided by the previously extracted median to generate a "daily variation characteristic index" consisting of multiple values. This index quantifies user electricity usage habits at different time periods, reflecting the degree to which peak and low electricity consumption periods deviate from normal levels. This daily variation characteristic index can be used to analyze user peak and valley characteristics of electricity consumption and help optimize power grid scheduling strategies.

[0074] 34. Extract the total amount of first historical electricity consumption data for multiple preset time periods corresponding to each of the multiple daily historical electricity consumption data;

[0075] For each preset time period (such as morning, midday, and evening peak hours), the corresponding total electricity consumption is extracted and the total electricity consumption data for multiple time periods is calculated. This is then used to analyze the electricity usage patterns of the same user in different time periods. By calculating the historical total electricity consumption for each preset time period, the user's electricity usage characteristics can be analyzed by time period.

[0076] 35. Calculate the second standard deviation between the total amount of multiple first historical electricity consumption data within the same preset time period;

[0077] 36. Use the second standard deviation as a cyclical indicator.

[0078] The standard deviation of total electricity consumption over multiple days within a preset time period is calculated as a periodicity indicator. The standard deviation reflects the consistency or cyclical variation of a user's electricity usage within the same time period. This metric can help identify regularities in user electricity usage behavior, such as whether there are cyclical fluctuations in electricity usage during the same time period each day of the week.

[0079] By integrating these indicators (volatility, median, diurnal variation, and periodicity), we can construct a comprehensive electricity consumption behavior analysis model. This model can be used to optimize scheduling and load management, such as load shifting and valley filling. These indicators can also be used to predict future user electricity demand, reducing power waste and power shortages during peak periods.

[0080] By calculating the first standard deviation of multiple historical electricity consumption data, a volatility index is obtained. This standard deviation reflects the degree of discreteness of the user's electricity consumption data, which can help identify the fluctuation range of electricity consumption, thereby providing support for the detection of abnormal fluctuations or unstable electricity consumption. Secondly, the first median of the historical electricity consumption data is extracted as the benchmark value of electricity consumption behavior, and the daily electricity consumption data is divided according to the preset time period. By calculating the ratio of the total amount of electricity consumption data in each time period to the median, a daily variation characteristic index is formed, which intuitively shows the difference in electricity consumption of users at different times of the day. This characteristic indicator can reveal the peaks and troughs of user electricity consumption and provide a basis for optimizing energy scheduling. Finally, in this application, the second standard deviation between the total amount of electricity consumption data in the same preset time period is calculated as a periodicity indicator. This indicator can effectively reveal the consistency and volatility of users in the periodic electricity consumption pattern, help to identify and analyze the periodic electricity consumption patterns of industrial and commercial users, and provide a more accurate reference for energy optimization and forecasting.

[0081] In summary, this application achieves a comprehensive analysis of users' historical electricity consumption data by extracting volatility, diurnal variation characteristics, and periodic indicators, and can provide more refined electricity management and optimization strategies for industrial and commercial users. According to the above technical solution, this application can achieve multi-dimensional characteristic analysis of industrial and commercial users' historical electricity consumption data, and by extracting volatility, diurnal variation characteristics, and periodic indicators, it provides an effective technical means for accurately characterizing users' electricity consumption behavior.

[0082] Optionally, when the load characteristics include a volatility index, a daily variation characteristic index, and a periodicity index, the data aggregation processing is performed based on the plurality of the aforementioned load characteristics to obtain a plurality of target aggregation sets, including:

[0083] 41. Extract the total amount of second historical electricity consumption data corresponding to multiple seasonal periods respectively;

[0084] 42. Calculate the third standard deviation between the total amount of the plurality of second historical electricity consumption data;

[0085] 43. The third standard deviation is used as the seasonality indicator.

[0086] Specifically, for the total electricity consumption data for multiple seasonal periods, the second historical electricity consumption data for these periods is extracted and the standard deviation between them is calculated to obtain the aforementioned third standard deviation as the seasonality indicator. This seasonality indicator is primarily used to measure the fluctuation characteristics of electricity consumption of industrial and commercial users in different seasons.

[0087] For example, we divide electricity consumption into time periods based on seasons (e.g., spring, summer, autumn, and winter), and calculate the total electricity consumption data for each season. For example, during the high temperatures of summer, equipment such as air conditioners may significantly increase electricity consumption. By calculating the standard deviation of this seasonal data, we can intuitively quantify the fluctuations in electricity consumption between seasons.

[0088] Seasonal indicators are suitable for predicting seasonal trends in electricity consumption peaks and troughs, and improving the reliability and adaptability of power supply.

[0089] The seasonality index is a relatively important characteristic dimension in load characteristics. To improve the calculation accuracy of the seasonality index, this embodiment also provides another method for calculating the seasonality index, which may include:

[0090] ① Calculate the trend indicators of historical electricity consumption data in the following way:

[0091]

[0092] Among them, T t represents the trend indicator of the tth period, m represents the number of seasons in a complete cycle (the number of seasons can be determined according to demand rather than set according to four fixed seasons), y t+i Represents the historical electricity consumption data of the t+i period.

[0093] ②Define the weight function:

[0094] W t =(y t -T t ) -2

[0095] Among them, y t represents the historical electricity consumption data of the tth period, W t represents the weight function.

[0096] ③Calculate the initial seasonal index:

[0097]

[0098] Among them, S i ′ represents the initial seasonal index of the i-th season, and s(t) is the season corresponding to the t-th period (i.e., s(t) = t mod m, if t mod ≠ m, otherwise s(t) = m)

[0099] ④ Normalize the initial seasonal index to obtain the final seasonal index:

[0100]

[0101] Among them, S i represents the final seasonal index of the i-th season (i.e., the seasonal index obtained in step 103). j ′ represents the initial seasonal index for the jth season. i and j are necessary parameters in the summation formula.

[0102] Combining the above steps, the calculation method of seasonal indicators is:

[0103]

[0104] 44. Extracting the time interval corresponding to the random event, wherein the random event refers to an event in which the multiple between the current power data and the power data on both sides is higher than a preset value;

[0105] 45. Calculate the proportion of the above time interval in the total duration and use the above proportion as the above randomness indicator.

[0106] In this embodiment, a random event is defined as the ratio of the current power data to the power data for the two time periods on either side exceeding a preset threshold. For example, a sudden surge in power load would be considered a random event. The time intervals of these random events are extracted to further analyze the user's power usage patterns.

[0107] Optionally, a multiple threshold (e.g., 2x, 3x, etc.) can be set. When the electricity usage data at a certain moment exceeds this multiple compared to the average value of the preceding and following time periods, it is considered that a random event exists at that moment. The length of these time intervals is extracted to assess the suddenness of user electricity usage behavior.

[0108] In this embodiment of the present application, the randomness index is defined as the ratio of the total duration of the random event time interval to the total observation time, thereby obtaining a numerical value reflecting the frequency of random events. This ratio index can intuitively reflect the distribution of random events in the user's electricity use over the entire electricity use time.

[0109] The proportion of random events can be calculated by calculating the ratio of the duration of a random event (e.g., minutes, hours, etc.) to the total duration (e.g., a week, a month, etc.). This metric reflects the stability of a user's electricity usage within a specific timeframe. The randomness index is suitable for assessing the stability and sporadic nature of electricity usage for industrial and commercial users. For example, in highly stable production processes, the randomness index should be low, while in enterprises that rely heavily on flexible production, the randomness index may be high.

[0110] In the embodiments of this application, seasonal and random indicators are combined to create a multi-dimensional profile of user electricity usage behavior. This profile helps to gain a deeper understanding of the user's electricity usage patterns and sudden characteristics over different time dimensions, providing a scientific basis for power demand side management, load forecasting, energy conservation and consumption reduction, etc.

[0111] By extracting the total amount of secondary historical electricity consumption data corresponding to multiple seasonal periods, calculating its third standard deviation, and using this standard deviation as a seasonality indicator, we can effectively capture the seasonal variations in historical electricity consumption data and provide an accurate measurement standard. This seasonality indicator facilitates forecasting electricity demand for future seasons and improves the stability and reliability of the power supply system.

[0112] At the same time, in the embodiment of the present application, random events are identified, the proportion of these events in the total duration is calculated, and this is used as a randomness index. These random events are abnormal situations where the multiple between the current power data and the power data on both sides is higher than the preset value. Through this method, the random change factors in the power consumption data are extracted, so that the system can better identify and respond to sudden power consumption fluctuations. This is of great significance for improving the level of power grid management, optimizing resource allocation, and enhancing the system's ability to resist risks.

[0113] In summary, the method in the embodiments of the present application can comprehensively analyze and quantify the seasonal and random factors in electricity consumption data, providing a scientific basis for the planning, monitoring and management of the power system, thereby improving the overall operating efficiency and emergency response capabilities.

[0114] 103. Calculate the similarity between the target cluster set and the preset load type.

[0115] The above-mentioned preset load types are pre-set load types, and different preset load types may correspond to different control strategies.

[0116] The above control strategies may be pre-set based on different characteristics corresponding to preset load types, according to data analysis and prior knowledge.

[0117] 104. If the similarity is greater than a first threshold, the control strategy corresponding to the preset load type is adopted to dispatch distributed power resources to multiple industrial and commercial users corresponding to the target cluster set.

[0118] The first threshold can be set as needed. If the calculated similarity is greater than the first threshold, the control strategy corresponding to the preset load type is adopted to dispatch distributed power resources to multiple industrial and commercial users corresponding to the target cluster set. Specifically, it may include:

[0119] Dispatching multiple distributed generation and / or energy storage systems to supply power to multiple industrial and commercial users corresponding to the target cluster set according to the control strategy corresponding to the preset load type;

[0120] The above-mentioned distributed generation may include wind power generation, photovoltaic power generation, hydropower generation and energy generation;

[0121] The above load characteristics include seasonality, periodicity, randomness and fluctuation;

[0122] The above preset load types include seasonal constant load, non-seasonal constant load, seasonal periodic load, non-seasonal periodic load, seasonal random load and non-seasonal random load.

[0123] Distributed generation includes, but is not limited to, combinations of one or more of wind power generation, photovoltaic power generation, hydropower generation, and energy generation. Different distributed generation systems exhibit one or more characteristics, such as volatility, periodicity, and seasonality. Load characteristics also exhibit one or more characteristics, such as volatility, periodicity, and seasonality. To adapt to these load characteristics, different distributed generation and energy storage systems need to be regulated to meet the electricity demands of different load characteristics. Preset load types include, but are not limited to, seasonal constant load, non-seasonal constant load, seasonal cyclical load, non-seasonal cyclical load, seasonal random load, and non-seasonal random load. For example, seasonal cyclical loads can be supplied by a combination of photovoltaic power generation (seasonality is primarily affected by high temperatures, and photovoltaic power generation is often high during high temperatures, so there is an approximately linear relationship between the two) and an energy storage system. Non-seasonal cyclical loads can be supplied by a combination of wind power generation (wind power generation is less affected by seasons), energy generation, and an energy storage system. Similarly, different regulation strategies can be determined based on actual application scenarios and needs, and are not limited here. The power supply of distributed generation and energy storage systems can be pre-set based on the historical total power consumption of aggregated industrial and commercial users and a preset redundancy.

[0124] In an optional implementation, after step 103, the following steps are further included:

[0125] If the similarity is not greater than the first threshold, the historical power consumption data corresponding to the target cluster set is sent to the power dispatching center, and the power dispatching center is used to set a control strategy corresponding to the historical power consumption data.

[0126] The power dispatching center is used to make settings based on historical power consumption data, using data analysis and prior knowledge to supplement new load characteristic data, thereby providing a more comprehensive data foundation for subsequent power regulation.

[0127] In an embodiment of the present application, historical electricity usage data for multiple industrial and commercial users is obtained to extract load characteristics, including volatility indicators, diurnal variation indicators, periodicity indicators, seasonality indicators, and randomness indicators, to comprehensively characterize the users' electricity usage behavior. Data aggregation is performed based on the extracted load characteristics to obtain multiple target cluster sets, enabling dynamic clustering of users with similar load characteristics and enhancing the flexibility and accuracy of load management. The similarity between the target cluster set and a preset load type is calculated. If the similarity is greater than a first threshold, the corresponding control strategy is adopted. This ensures that the control measures are highly consistent with the users' actual electricity usage characteristics, avoiding the inefficiency caused by user differences in traditional control. Dispatching distributed power resources to the multiple industrial and commercial users corresponding to the target cluster set improves the utilization efficiency of distributed energy, reduces peak-to-valley differences in the power grid, and alleviates power supply pressure. In summary, the method in the embodiment of the present application, through refined analysis and dynamic aggregation of the load characteristics of industrial and commercial users, can achieve precise matching of control strategies and efficient scheduling of power resources, significantly improving the rationality of load control and the utilization efficiency of power resources, and ensuring the safe and stable operation of the power grid.

[0128] The following is a case analysis based on the description of the above method embodiment and combined with specific scenarios.

[0129] Taking a certain industrial park as an example, the rationality and application value of the virtual power plant control strategy are analyzed based on the control model. Figure 2Figure 2 shows a typical daily load and distributed photovoltaic energy curve for a certain industrial park. It can be seen that the industrial park load exhibits a distinct bimodal characteristic, and the distributed photovoltaic energy within the region provides a strong output during the day. However, even when fully utilizing distributed photovoltaic energy, the regional grid's residual load exhibits significant peak-to-valley variations and volatility. Leveraging the advantages of virtual power plants, mobilizing regional demand response resources, and guiding electricity users' energy needs can reduce grid load volatility, improve power quality, and lower grid peak and frequency regulation costs. Assume that multiple virtual power plants within a region possess different response resources, resulting in varying response costs. However, from a grid dispatch perspective, based on an assessment of virtual power plant response capabilities, those with the lowest response costs will be prioritized. Competition among virtual power plants and the adjustment of response resources will ultimately lead to a more balanced response cost. Therefore, this article does not discuss the case of virtual power plants with varying resource allocations. Instead, it focuses on the impact of varying demand response resource ratios on grid operation from the perspective of grid dispatch and operation.

[0130] Figure 3 The results for a regional virtual power plant with a total of 2MW of energy storage resources (including electric vehicles) and several different adjustable load ratios (relative to the grid load) are shown. This shows that distributed energy resources are fully utilized during regulation. Given limited energy storage resources, a low adjustable load ratio has limited impact on the overall load curve. However, increasing the adjustable load ratio significantly improves peak load regulation. This demonstrates the importance of fully leveraging the adjustable load in virtual power plant construction.

[0131] Figure 4 The results of regulation when the regional virtual power plant contains a total of 4MW of energy storage resources (including electric vehicles) and several different adjustable load ratios are shown. It can be seen that under the premise of making full use of distributed energy, when the energy storage resources in the virtual power plant are increased, it can also have good peak regulation capabilities when the adjustable load ratio is small. As the adjustable load ratio increases, the peak and valley regulation capabilities of the load curve are further enhanced. By comparison Figure 3 、 4 As can be seen, energy storage and adjustable loads are both important resources in virtual power plants. Increasing energy storage resources and directing adjustable loads to participate in system regulation can optimize resource utilization, promote the absorption of new energy, and improve the power system's regulatory capacity. However, regulating both energy storage and adjustable loads carries associated costs. Therefore, based on the results of this virtual power plant regulation strategy, based on demand response capability assessment and the classification and aggregation of virtual power plants, it is possible to selectively regulate virtual power plants with significant resource advantages and low costs to participate in system regulation, thereby achieving the goal of zoning regulation and resource optimization.

[0132] Based on the description of the aforementioned method embodiment, the embodiment of the present application also provides an aggregator-level dynamic aggregation and control device that is adaptable to industrial and commercial loads.

[0133] Figure 5 This is a schematic diagram of the structure of a dynamic aggregation and control device for industrial and commercial loads provided by an embodiment of the present application. Figure 5 As shown, the aggregator-level dynamic aggregation and control device 500 adapted to industrial and commercial loads includes:

[0134] The data acquisition module 510 is used to acquire historical electricity consumption data of multiple industrial and commercial users;

[0135] A characteristic extraction module 520 is used to extract load characteristics of the historical power consumption data;

[0136] an aggregation processing module 530, configured to perform data aggregation processing according to the plurality of load characteristics to obtain a plurality of target aggregation sets;

[0137] A similarity calculation module 540 is used to calculate the similarity between the target cluster set and the preset load type;

[0138] The control strategy execution module 550 is configured to, if the similarity is greater than a first threshold, adopt the control strategy corresponding to the preset load type to dispatch distributed power resources to multiple industrial and commercial users corresponding to the target cluster set.

[0139] Understandably, Figure 5 The relevant contents of each module in the above method embodiment have been described in detail, and the details can be referred to the contents of the method embodiment; Figure 5 The provided aggregator-level dynamic aggregation and control device 500 adapted to industrial and commercial loads can perform the following operations: Figure 1 Any steps in the illustrated embodiment will not be described in detail here.

[0140] The present application provides an aggregator-level dynamic aggregation and control device 200 that is suitable for industrial and commercial loads. By acquiring historical electricity consumption data of multiple industrial and commercial users, it can extract load characteristics including volatility indicators, daily variation characteristic indicators, periodicity indicators, seasonality indicators and randomness indicators, and comprehensively characterize the user's electricity consumption behavior. Based on the extracted load characteristics, data aggregation processing is performed to obtain multiple target aggregation sets, thereby realizing dynamic clustering of users with similar load characteristics and enhancing the flexibility and accuracy of load management. The similarity between the target cluster set and the preset load type is calculated. If the similarity is greater than the first threshold, the corresponding control strategy is adopted. This ensures that the control measures are highly matched with the actual electricity consumption characteristics of the users, avoiding the inefficiency problem caused by user differences in traditional control. Dispatching distributed power resources to multiple industrial and commercial users corresponding to the target cluster set improves the utilization efficiency of distributed energy, reduces the peak-to-valley difference of the power grid, and alleviates the pressure on power supply. To sum up, this application achieves accurate matching of control strategies and efficient scheduling of power resources through refined analysis and dynamic aggregation of the load characteristics of industrial and commercial users, significantly improves the rationality of load control and the utilization efficiency of power resources, and ensures the safe and stable operation of the power grid.

[0141] In one embodiment of the present application, an electronic device is also provided. Figure 6 , Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 6 As shown, the electronic device 600 includes a processor 601 and a memory 602. The memory 602 stores a computer program. When the computer program is executed by the processor 601, the following operations are performed: Figure 1 Any step in the method embodiment shown; or it can be understood that the processor 601 implements the functions of each module in the above-mentioned device embodiments when executing the computer program, such as Figure 5 The electronic device 600 may also include input / output devices, etc. In a specific embodiment, the electronic device may be a terminal device, etc.

[0142] In one embodiment, a computer-readable storage medium is further provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor is caused to perform any step in the above method embodiment.

[0143] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. 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 in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various 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).

[0144] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0145] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for dynamic aggregation and control of aggregators at the aggregator level adapted to industrial and commercial loads, characterized in that: The method comprises: Obtain historical electricity consumption data for multiple industrial and commercial users; Extracting load characteristics of the historical electricity consumption data, performing data aggregation processing based on a plurality of the load characteristics, and obtaining a plurality of target aggregation sets; Calculating the similarity between the target cluster set and the preset load type; If the similarity is greater than a first threshold, the control strategy corresponding to the preset load type is adopted to dispatch distributed power resources to multiple industrial and commercial users corresponding to the target cluster set.

2. The method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads according to claim 1, characterized in that: The load characteristics include a volatility index, a daily variation characteristic index and a periodicity index; The extracting load characteristics of the historical electricity usage data includes: For each industrial and commercial user, calculate a first standard deviation of the plurality of historical electricity usage data as the volatility index; Extracting a first median of a plurality of historical electricity usage data corresponding to each industrial and commercial user; Dividing the daily historical electricity consumption data into preset time periods, dividing the total amount of first historical electricity consumption data corresponding to each of the multiple preset time periods by the first median, and obtaining the daily variation characteristic index consisting of multiple values; The total amount of first historical electricity consumption data of multiple preset time periods corresponding to multiple daily historical electricity consumption data is extracted respectively, the second standard deviation between the multiple total amounts of first historical electricity consumption data in the same preset time period is calculated, and the second standard deviation is used as the periodicity indicator.

3. The method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads according to claim 1, characterized in that: The load characteristics include seasonality index and randomness index; The extracting load characteristics of the historical electricity usage data includes: extracting the total amount of second historical electricity consumption data corresponding to multiple seasonal periods respectively, calculating the third standard deviation between the multiple total amounts of second historical electricity consumption data, and using the third standard deviation as the seasonality indicator; Extracting the time interval corresponding to the random event, wherein the random event refers to an event in which the multiple between the current power data and the power data on both sides is higher than a preset value; The proportion of the time interval in the total duration is calculated, and the proportion is used as the randomness index.

4. The method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads according to claim 3 is characterized in that: Performing data aggregation processing according to the plurality of load characteristics to obtain a plurality of target aggregation sets includes: Encoding the multiple load characteristics to obtain multiple code values, and constructing the multiple code values into a load characteristic vector in a preset order; Randomly extracting a preset number of load characteristic vectors, and calculating distances between the preset number of load characteristic vectors; The load characteristic vectors whose distance is less than the second threshold are taken as the same initial aggregation set, and the load characteristic vectors whose distance is not less than the second threshold are taken as two initial aggregation sets respectively; The distances between other load characteristic vectors and multiple initial aggregation sets are calculated in sequence; Incorporating other load characteristic vectors whose distance is less than the second threshold into the initial aggregate set, and taking other load characteristic vectors whose distance is not less than the second threshold as a new initial aggregate set; The multiple initial aggregation sets finally obtained are used as the multiple target aggregation sets.

5. The method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads according to claim 4 is characterized in that: The preset load types include seasonal constant load, non-seasonal constant load, seasonal periodic load, non-seasonal periodic load, seasonal random load and non-seasonal random load; The distributed power generation includes wind power generation, photovoltaic power generation, hydropower generation and energy power generation; The control strategy is pre-set based on the load characteristics corresponding to the preset load type, and the load characteristics include seasonality, periodicity, randomness and volatility.

6. The method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads according to claim 4, characterized in that: The method further comprises: If the similarity is not greater than the first threshold, the historical power consumption data corresponding to the target cluster set is sent to a power dispatching center, and the power dispatching center is used to set a control strategy corresponding to the historical power consumption data.

7. The method for dynamic aggregation and control at the aggregator level adapted to industrial and commercial loads according to claim 1, characterized in that: The method further comprises: Based on the real-time electricity consumption data of multiple industrial and commercial users corresponding to the target cluster set, the scheduling of the distributed power resources is monitored and adjusted in real time; and / or, The control strategy is optimized and adjusted according to the electricity consumption feedback information of multiple industrial and commercial users corresponding to the target cluster set.

8. A dynamic aggregation and control device at the aggregator level adapted to industrial and commercial loads, characterized in that: include: A data acquisition module is used to obtain historical electricity consumption data of multiple industrial and commercial users; A characteristic extraction module, configured to extract load characteristics of the historical power consumption data; an aggregation processing module, configured to perform data aggregation processing according to the plurality of load characteristics to obtain a plurality of target aggregation sets; A similarity calculation module, configured to calculate the similarity between the target cluster set and a preset load type; The control strategy execution module is used to adopt the control strategy corresponding to the preset load type to dispatch distributed power resources to multiple industrial and commercial users corresponding to the target cluster set if the similarity is greater than a first threshold.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

Images