A controllable load clustering method and system for supporting balanced computation

Abstract

This invention discloses a controllable load clustering method and system supporting load balancing calculations. The method includes: processing typical controllable loads to establish a physical model of the controllable loads describing their electricity consumption characteristics; considering the time-related usage needs of electricity users for the physical model of the controllable loads, abstracting the electricity consumption characteristics of the controllable loads, and establishing multiple types of controllable load models that take into account these characteristics; and establishing... l Electricity consumption characteristic data is used as clustering objects; k-means The algorithm is suitable for massive controllable loads. l This system performs cluster analysis on electricity consumption characteristic data; based on the clustered controllable load categories, it analyzes the conditions that controllable loads with interactive relationships must meet to reach a balanced state, which are then used for subsequent balance calculations. The system includes a model building module, a model processing module, a cluster analysis module, and a condition analysis module. This invention lays the foundation for the participation of massive amounts of controllable loads in demand-side management.

Description

Technical Field

[0001] This invention belongs to the field of power systems, and specifically relates to a controllable load clustering method and system that supports load balancing calculation. Background Technology

[0002] Building a new power system under the backdrop of energy transition faces challenges. The increasing proportion of renewable energy and the rapid growth of new loads such as electric vehicles have further exacerbated the daily fluctuations of renewable energy and widened the peak-to-valley load gap. However, conventional power sources have limited regulation capabilities, making it difficult to ensure the safe and stable operation of the power system. Therefore, it is necessary to explore the regulation potential of load-side resources and enable users to participate deeply in system balancing. Residential loads are large in scale, growing rapidly, and highly intelligent, serving as an important supplement to demand response dominated by large industrial and commercial loads, and a crucial foundation for supporting supply and demand interaction.

[0003] However, residential controllable loads are characterized by small individual capacity, large quantity, diverse types, and dispersed distribution, leading to random response issues and making it difficult for dispatch centers to obtain their power consumption. Therefore, residential loads cannot directly participate in demand-side management. With the development of smart terminals and communication technologies for power users, the power system possesses massive amounts of basic user electricity consumption data. Therefore, most existing studies employ data mining to cluster power load curves, starting with overall user electricity consumption data, either directly clustering user load curves or indirectly using data dimensionality reduction techniques and time series analysis to extract key features of load curves, further clustering analysis to reflect the electricity consumption behavior characteristics of similar users and achieve power user classification. However, using statistical methods to analyze overall user electricity consumption data, extending spatially and considering user electricity consumption cluster behavior, cannot guarantee physical compliance with controllable load operation when implementing demand response decisions for residential controllable loads. To ensure the feasibility of the control layer's participation in demand response, a controllable load clustering method supporting load balancing calculation is proposed. Taking controllable load terminal equipment as the research object, a mathematical model of the power consumption characteristics of typical controllable terminal equipment is constructed. Furthermore, based on the characteristics of equipment operation and user demand, the controllable features of terminal equipment are abstracted to form a multi-type controllable load model that takes into account power consumption characteristics. Cluster analysis is then performed on massive, multi-type controllable loads. Finally, based on the clustering results obtained from load characteristics, the equilibrium state ultimately reached by controllable loads of the same type in the interaction is analyzed, and category-based equilibrium conditions are constructed to provide a foundation for subsequent calculation of this equilibrium state.

[0004] In summary, cluster analysis of power loads is the foundation for demand-side management and energy efficiency management in future new power systems. However, most existing research focuses on data mining, which is often disconnected from the control layer of controllable loads and cannot determine the electricity consumption behavior of the underlying controllable loads in demand response. Summary of the Invention

[0005] The purpose of this invention is to address the problem of the disconnect between existing clustering methods and the controllable load control layer. Starting from the physical operating characteristics of typical controllable loads, a mathematical model describing their power consumption characteristics is constructed. This model further abstracts the controllable features of a large number of devices, forming multi-type controllable load models. This provides a controllable load clustering method and system that supports balancing calculations for efficient modeling of massive controllable load resources. It describes the power consumption characteristics of controllable loads of the same type, laying the foundation for the participation of massive controllable loads at the bottom layer in demand-side management.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A controllable load clustering method supporting load balancing calculations includes:

[0008] Typical controllable loads are processed to establish a physical model of controllable loads that describes their power consumption characteristics;

[0009] Considering the time-related usage requirements of power users for the physical model of controllable load, the electricity consumption characteristics of controllable load are abstracted, and multiple types of controllable load models that take into account the electricity consumption characteristics are established.

[0010] For multiple types of controllable load models that take into account electricity consumption characteristics, establish l Electricity consumption characteristic data is used as clustering objects;

[0011] use k-means The algorithm is suitable for massive controllable loads. l Cluster analysis was performed on the electricity consumption characteristic data.

[0012] Based on the controllable loads obtained from clustering, we analyze the conditions that controllable loads of the same type with interactive relationships must satisfy to reach equilibrium, which are then used for subsequent equilibrium calculations.

[0013] A further improvement of this invention lies in processing typical controllable loads and establishing a physical model of the controllable load describing its electrical characteristics, including:

[0014] Electricity users have multiple controllable loads. For electric vehicles, factors such as user arrival time, departure time, charging power limits, and total electricity consumption need to be considered.

[0015] (1)

[0016] in, H Indicates the entire calculation cycle; H e This is the permissible charging time for electric vehicles; outside this range, the charging power of the electric vehicle is 0. p e,h Indicates electric vehicles e existh Energy consumption over time, p e,min , p e,max [This refers to the charging power range for electric vehicles.] p e,min This indicates the minimum charging power of an electric vehicle. p e,max Indicates the maximum charging power of an electric vehicle; E e,min , E e,max [This refers to the range of total electricity consumption for electric vehicles.] E e,min This indicates the minimum electricity consumption of an electric vehicle. E e,max This indicates the maximum power consumption of an electric vehicle.

[0017] A further improvement of this invention lies in considering the time-related usage requirements of power users for the physical model of controllable loads, abstracting the electricity consumption characteristics of controllable loads, and establishing multiple types of controllable load models that take into account these electricity consumption characteristics, including:

[0018] (2)

[0019] (3)

[0020] (4)

[0021] in, Indicates electric vehicles e In time h The state variable represents whether the electric vehicle is allowed to be charged;

[0022] Further abstracting the electricity consumption characteristics of electric vehicles yields electricity consumption characteristic data to be clustered.

[0023] (5)

[0024] in For electric vehicles e State variables throughout the entire computation cycle; For electric vehicles e The minimum power state sequence; For electric vehicles e The maximum power state sequence.

[0025] A further improvement of this invention lies in establishing a model for multiple types of controllable loads that take into account electricity consumption characteristics. l Electricity consumption characteristic data, as clustering objects, include:

[0026] (6)

[0027] Using a partition-based clustering algorithm k-means The algorithm is suitable for massive controllable loads. l Cluster analysis was performed on the electricity consumption characteristic data, and different load clusters were selected based on the size of the controllable load. K , K <N Find a partition ( S k ) 1≤k≤K Divide the load into K cluster

[0028] (7)

[0029] in Represents load clusters S k The cluster centers, i.e., the clusters after clustering. k Electrical consumption characteristics of similar loads ; N The number of controllable loads.

[0030] A further improvement of this invention lies in analyzing the conditions that controllable loads of the same category with interactive relationships must satisfy to reach equilibrium, based on the controllable loads obtained from clustering, for subsequent equilibrium calculations, including:

[0031] (8)

[0032] (9)

[0033] in For indicator functions, when hour, ,otherwise ; This represents the feasible operating domain of the load generated by the electricity consumption characteristic mapping, meaning that each controllable load must meet the requirements of its category. k The feasible operating domain; This represents the electricity price function derived from the mapping of load behavior. Indicates controllable load a Electricity costs;

[0034] When equilibrium exists and is reached, the electricity cost of each load is minimized, meaning no load can further optimize its outcome by changing its own electricity consumption strategy, satisfying the following equilibrium condition.

[0035] (10)

[0036] (11)

[0037] (12)

[0038] in Indicates controllable load a The optimal power consumption scheme under balanced conditions; yes k The set of optimal power consumption schemes under controllable load balancing conditions is a finite set. yes k Select the first type of controllable load j The electricity cost of various electricity usage plans; express k Select the first type of controllable load j The number of different electricity usage schemes; express k The number of controllable loads.

[0039] A controllable load clustering system supporting load balancing calculations includes:

[0040] The first model building module processes typical controllable loads and establishes a physical model of the controllable load that describes the characteristics of electricity consumption.

[0041] The second model building module considers the time-related usage requirements of power users for the physical model of controllable loads, abstracts the electricity consumption characteristics of controllable loads, and establishes multiple types of controllable load models that take into account the electricity consumption characteristics.

[0042] The model processing module establishes models for various types of controllable loads that take into account electricity consumption characteristics. l Electricity consumption characteristic data is used as clustering objects;

[0043] Cluster analysis module, using k-means The algorithm is suitable for massive controllable loads. l Cluster analysis was performed on the electricity consumption characteristic data.

[0044] The condition analysis module analyzes the conditions that controllable loads of the same type with interactive relationships must meet to reach equilibrium, based on the controllable loads obtained from clustering, and uses this analysis for subsequent equilibrium calculations.

[0045] A further improvement of this invention lies in processing typical controllable loads and establishing a physical model of the controllable load describing its electrical characteristics, including:

[0046] Electricity users have multiple controllable loads. For electric vehicles, factors such as user arrival time, departure time, charging power limits, and total electricity consumption need to be considered.

[0047] (1)

[0048] in, H Indicates the entire calculation cycle;H e This is the permissible charging time for electric vehicles; outside this range, the charging power of the electric vehicle is 0. p e,h Indicates electric vehicles e exist h Energy consumption over time, p e,min , p e,max [This refers to the charging power range for electric vehicles.] p e,min This indicates the minimum charging power of an electric vehicle. p e,max Indicates the maximum charging power of an electric vehicle; E e,min , E e,max [This refers to the range of total electricity consumption for electric vehicles.] E e,min This indicates the minimum electricity consumption of an electric vehicle. E e,max This indicates the maximum power consumption of an electric vehicle.

[0049] A further improvement of this invention lies in considering the time-related usage requirements of power users for the physical model of controllable loads, abstracting the electricity consumption characteristics of controllable loads, and establishing multiple types of controllable load models that take into account these electricity consumption characteristics, including:

[0050] (2)

[0051] (3)

[0052] (4)

[0053] in, Indicates electric vehicles e In time h The state variable represents whether the electric vehicle is allowed to be charged;

[0054] Further abstracting the electricity consumption characteristics of electric vehicles yields electricity consumption characteristic data to be clustered.

[0055] (5)

[0056] in For electric vehicles e State variables throughout the entire computation cycle; For electric vehicles e The minimum power state sequence; For electric vehicles e The maximum power state sequence.

[0057] A further improvement of this invention lies in establishing a model for multiple types of controllable loads that take into account electricity consumption characteristics. l Electricity consumption characteristic data, as clustering objects, include:

[0058] (6)

[0059] Using a partition-based clustering algorithm k-means The algorithm is suitable for massive controllable loads. l Cluster analysis was performed on the electricity consumption characteristic data, and different load clusters were selected based on the size of the controllable load. K , K <N Find a partition ( S k ) 1≤k≤K Divide the load into K cluster

[0060] (7)

[0061] in Represents load clusters S k The cluster centers, i.e., the clusters after clustering. k Electrical consumption characteristics of similar loads ; N The number of controllable loads.

[0062] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a controllable load clustering method supporting balanced computation.

[0063] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0064] This invention starts from the operating characteristics of controllable loads and user needs. First, it establishes a physical model of the controllable load describing its operating characteristics by considering the physical characteristics of the load operation. Next, it considers the user needs of the controllable load and establishes multiple types of controllable load models. Finally, it extracts electricity consumption characteristic data as clustering objects and adopts... k-means The algorithm performs cluster analysis on multiple types of controllable loads, and the clustering results can be used to construct category-based equilibrium conditions. Compared with existing traditional load clustering methods based on large datasets, this invention solves the problems of traditional controllable load clustering methods being disconnected from the control layer of controllable loads, failing to consider the operating characteristics of controllable loads, and having high computational complexity. It provides an effective quantitative reference for the next step of calculating the equilibrium state of massive controllable loads considering the impact of load response on electricity prices. Attached Figure Description

[0065] Figure 1 This is a flowchart of a controllable load clustering method that supports load balancing calculation according to the present invention;

[0066] Figure 2 A graph showing the relationship between effectiveness metrics and the number of clusters;

[0067] Figure 3 This is a structural block diagram of a controllable load clustering system that supports balanced calculation according to the present invention. Detailed Implementation

[0068] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0069] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0070] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0071] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0072] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0073] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0074] Example 1

[0075] like Figure 1As shown, the present invention provides a controllable load clustering method for supporting load balancing calculations, comprising:

[0076] Step 1: Process typical controllable loads and establish a physical model of the controllable loads that describes their power consumption characteristics;

[0077] Step 2: Considering the time-related usage requirements of power users for the physical model of controllable loads, abstract the electricity consumption characteristics of controllable loads and establish multi-type controllable load models that take into account the electricity consumption characteristics;

[0078] Step 3: Establish a multi-type controllable load model that takes into account electricity consumption characteristics. l Electricity consumption characteristic data is used as clustering objects;

[0079] Step 4: Use k-means The algorithm is suitable for massive controllable loads. l Cluster analysis was performed on the electricity consumption characteristic data.

[0080] Step 5: Based on the controllable loads obtained from clustering, analyze the conditions that controllable loads of the same type with interactive relationships must meet to reach equilibrium, which will be used for subsequent equilibrium calculations.

[0081] In this embodiment, in step one, the electricity user has multiple controllable loads. For electric vehicles, the user's arrival time, departure time, charging power upper and lower limits, and total electricity consumption are considered.

[0082] (1)

[0083] in, H Indicates the entire calculation cycle; H e This is the permissible charging time for electric vehicles; outside this range, the charging power of the electric vehicle is 0. p e,h Indicates electric vehicles e exist h Energy consumption over time, p e,min , p e,max [This refers to the charging power range for electric vehicles.] p e,min This indicates the minimum charging power of an electric vehicle. p e,max Indicates the maximum charging power of an electric vehicle; E e,min , E e,max [This refers to the range of total electricity consumption for electric vehicles.] E e,min This indicates the minimum electricity consumption of an electric vehicle. E e,maxThis indicates the maximum power consumption of an electric vehicle.

[0084] In this embodiment, step two, taking an electric vehicle as an example, includes:

[0085] (2)

[0086] (3)

[0087] (4)

[0088] in, Indicates electric vehicles e In time h The state variable represents whether the electric vehicle is allowed to be charged;

[0089] Further abstracting the electricity consumption characteristics of electric vehicles yields electricity consumption characteristic data to be clustered.

[0090] (5)

[0091] in For electric vehicles e State variables throughout the entire computation cycle; For electric vehicles e The minimum power state sequence; For electric vehicles e The maximum power state sequence.

[0092] Step three specifically includes:

[0093] (6)

[0094] Using a partition-based clustering algorithm k-means The algorithm is suitable for massive controllable loads. l Cluster analysis was performed on the electricity consumption characteristic data, and different load clusters were selected based on the size of the controllable load. K ( K <N ), find a partition ( S k ) 1≤k≤K Divide the load into K cluster

[0095] (7)

[0096] in Represents load clusters S k The cluster centers, i.e., the clusters after clustering. k Electrical consumption characteristics of similar loads .

[0097] Step 5: Specifically includes:

[0098] Based on the clustering results, a category-based equilibrium condition is constructed. Considering the impact of load response on electricity prices, controllable loads engage in a game theory relationship to minimize their own electricity costs.

[0099] (8)

[0100] (9)

[0101] in For indicator functions, when hour, ,otherwise ; This represents the feasible operating domain of the load generated by the electricity consumption characteristic mapping, meaning that each controllable load must meet the requirements of its category. k The feasible operating domain; This represents the electricity price function derived from the mapping of load behavior. Indicates controllable load a Electricity costs.

[0102] When equilibrium exists and is reached, the electricity cost of each load is minimized, meaning no load can further optimize its outcome by changing its own electricity consumption strategy, satisfying the following equilibrium condition.

[0103] (10)

[0104] (11)

[0105] (12)

[0106] in Indicates controllable load a The optimal power consumption scheme under balanced conditions; yes k The set of optimal power consumption schemes under controllable load balancing conditions is a finite set. yes k Select the first type of controllable load j The electricity cost of various electricity usage plans; express k Select the first type of controllable load j The number of different electricity usage schemes; express k The number of controllable loads.

[0107] Example 2

[0108] Obtain controllable load data to be clustered N One, obtained through preprocessingl The data on electricity consumption characteristics includes preprocessing, which specifically describes the physical characteristics of controllable loads and establishes multiple types of controllable load models based on load usage requirements. The following steps are then performed. k-means Clustering steps: Set the number of clusters n And convergence criteria, and randomly select n indivual l The electricity consumption characteristic data is used as the cluster center, denoted as .set up t = 0, 1, 2, … represents the number of iterations. Repeat the following steps until the convergence criterion is met: 1) For each l Based on the electricity consumption characteristic data, calculate the Euclidean distance to all cluster centers and assign the data to the nearest center. ,in Indicates the first i The class to which each sample belongs, each The value range is 1 to n ;2) For each class n Recalculate the cluster centers of this class. .

[0109] The effectiveness of clustering methods is assessed using the sum of squared errors (SSE). SSE This represents the sum of the Euclidean distances from all samples to the cluster centers of their respective classes:

[0110] (13)

[0111] As the number of clusters increases, SSE It will gradually decrease, especially after the inflection point. SSE The rate of decrease becomes smaller, so the inflection point is chosen as the optimal number of clusters.

[0112] Considering 2000 controllable load data points, the number of clusters is set from 5 to 25. SSE Indicators such as Figure 1 As shown. By Figure 2 It can be seen that the number of clusters n =19 is the inflection point of the SSE curve, therefore the optimal number of clusters is 19. Furthermore, a category-based equilibrium condition can be constructed based on the clustering results.

[0113] Example 3

[0114] like Figure 3 As shown, the present invention provides a controllable load clustering system that supports load balancing calculations, comprising:

[0115] The first model building module processes typical controllable loads and establishes a physical model of the controllable load that describes the characteristics of electricity consumption.

[0116] The second model building module considers the time-related usage requirements of power users for the physical model of controllable loads, abstracts the electricity consumption characteristics of controllable loads, and establishes multiple types of controllable load models that take into account the electricity consumption characteristics.

[0117] The model processing module establishes models for various types of controllable loads that take into account electricity consumption characteristics. l Electricity consumption characteristic data is used as clustering objects;

[0118] Cluster analysis module, using k-means The algorithm is suitable for massive controllable loads. l Cluster analysis was performed on the electricity consumption characteristic data.

[0119] The condition analysis module analyzes the conditions that controllable loads of the same type with interactive relationships must meet to reach equilibrium, based on the controllable loads obtained from clustering, and uses this analysis for subsequent equilibrium calculations.

[0120] Example 4

[0121] The present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a controllable load clustering method supporting balanced computation.

[0122] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0123] This application is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0124] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0125] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0126] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0127] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A controllable load clustering method to support balanced computation, characterized in that, include: Typical controllable loads are processed, and a physical model of the controllable load describing its power consumption characteristics is established, including: Electricity users have multiple controllable loads. For electric vehicles, factors such as user arrival time, departure time, charging power limits, and total electricity consumption need to be considered. where H denotes the total computation period; H e is the allowed charging time of the electric vehicle, outside of which the electric vehicle charging power is 0; p e,h denotes the electric energy consumption of the electric vehicle e at time h, [p e,min ,p e,max ] is the electric vehicle charging power interval, p e,min denotes the minimum charging power of the electric vehicle, p e,max denotes the maximum charging power of the electric vehicle; [E e,min ,E e,max ] is the electric vehicle total electric energy consumption interval, E e,min denotes the minimum electric energy consumption of the electric vehicle, E e,max denotes the maximum electric energy consumption of the electric vehicle; Considering the time-related usage requirements of power users for the physical model of controllable loads, the electricity consumption characteristics of controllable loads are abstracted, and multiple types of controllable load models that take into account these characteristics are established, including: wherein X e,h represents the state variable of the electric vehicle e at time h, representing whether the electric vehicle is allowed to charge; Further abstracting the electricity consumption characteristics of electric vehicles yields electricity consumption characteristic data to be clustered. v e = (p e,min X e , p e,max X e , E e,min , E e,max )(5) where X e is the state variable of the electric vehicle e over the entire calculation period; p e,min X e is the minimum power state sequence of the electric vehicle e; p e,max X e is the maximum power state sequence of the electric vehicle e; For multi-type controllable load models that take into account electricity consumption characteristics, an l-dimensional electricity consumption characteristic data is established as a clustering object, including: v a = (p a,min X a , p a,max X a , E a,min , E a,max ) ∈ R l (6) The k-means algorithm based on division is used for clustering analysis of the 1-dimensional power consumption characteristic data of the massive controllable load, different load clusters K are selected according to the size of the controllable load, K k ) 1≤k≤K The load is divided into K clusters wherein represents the load cluster S k the clustering center of the kth cluster, i.e., the power consumption characteristics of the kth cluster after clustering N is the number of controllable loads; The k-means algorithm was used to perform cluster analysis on the l-dimensional electricity consumption characteristic data of massive controllable loads; Based on the controllable loads obtained from clustering, we analyze the conditions that controllable loads of the same type with interactive relationships must satisfy to reach equilibrium, which are then used for subsequent equilibrium calculations.

2. The controllable load clustering method for supporting equalization computation according to claim 1, wherein, Based on the controllable load categories obtained from clustering, analyze the conditions that controllable loads of the same category with interactive relationships must satisfy to reach equilibrium, which are used for subsequent equilibrium calculations, including: where is an indicator function, when a ∈ S k , otherwise f(v k ) represents the feasible operating region of the load generated by the load feature map, i.e., each controllable load must satisfy the feasible operating region of its belonging class k; λ(·) represents the electricity price function derived from the load behavior map; c a represents the electricity cost of the controllable load a. When equilibrium exists and is reached, the electricity cost of each load is minimized, meaning no load further optimizes its outcome by changing its own electricity consumption strategy, satisfying the following equilibrium condition. wherein represents the optimal power consumption scheme in the balanced state of controllable load a; Q k is the optimal power consumption scheme set in the balanced state of k controllable loads, and is a finite set; c k,j is the power consumption cost of the jth power consumption scheme selected by the k controllable loads; n k,j represents the number of the jth power consumption scheme selected by the k controllable loads; |S k | represents the number of the k controllable loads.

3. A controllable load clustering system supporting balanced computation, characterized by, include: The first model building module processes typical controllable loads and establishes a physical model of the controllable load describing its power consumption characteristics, including: Electricity users have multiple controllable loads. For electric vehicles, factors such as user arrival time, departure time, charging power limits, and total electricity consumption need to be considered. where H denotes the total computation period; H e is the allowed charging time of the electric vehicle, outside of which the electric vehicle charging power is 0; p e,h denotes the electric energy consumption of the electric vehicle e at time h, [p e,min , p e,max ] is the electric vehicle charging power interval, p e,min denotes the minimum charging power of the electric vehicle, p e,max denotes the maximum charging power of the electric vehicle; [E e,min , E e,max ] is the electric vehicle total electric energy consumption interval, E e,min denotes the minimum electric energy consumption of the electric vehicle, E e,max denotes the maximum electric energy consumption of the electric vehicle; The second model building module considers the time-related usage requirements of electricity users for the physical model of controllable loads, abstracts the electricity consumption characteristics of controllable loads, and establishes multiple types of controllable load models that take into account these characteristics, including: wherein X e,h represents the state variable of the electric vehicle e at time h, representing whether the electric vehicle is allowed to charge; Further abstracting the electricity consumption characteristics of electric vehicles yields electricity consumption characteristic data to be clustered. v e = (p e,min X e ,p e,max X e ,E e,min ,E e,max )(5) where X e is the state variable of the electric vehicle e over the entire calculation period; p e,min X e is the minimum power state sequence of the electric vehicle e; p e,max X e is the maximum power state sequence of the electric vehicle e; The model processing module, for multiple types of controllable load models that take into account electricity consumption characteristics, establishes l-dimensional electricity consumption characteristic data as clustering objects, including: v a = (p a,min X a , p a,max X a , E a,min , E a,max ) ∈ R l (6) The k-means algorithm based on division is used for clustering analysis of the 1-dimensional power consumption characteristic data of the massive controllable load, different load clusters K are selected according to the size of the controllable load, K k ) 1≤k≤K The load is divided into K clusters wherein represents the cluster center of the load cluster S k , i.e. the electricity consumption feature of the kth cluster after clustering N is the number of controllable loads; The clustering analysis module uses the k-means algorithm to perform clustering analysis on the l-dimensional electricity consumption characteristic data of massive controllable loads; The condition analysis module analyzes the conditions that controllable loads of the same type with interactive relationships must meet to reach equilibrium, based on the controllable loads obtained from clustering, and uses this analysis for subsequent equilibrium calculations.

4. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a controllable load clustering method supporting balanced computation as described in claim 1 or 2.

Images