Virtual power plant unit dynamic division method and system based on power grid topology

Abstract

The invention discloses a dynamic division method and system for virtual power plant units based on power grid topology, and the method comprises the steps: mapping adjustable resources of a virtual power plant to power grid control nodes, forming a physical link relation database, and building a resource-power grid topology mapping model; setting a multi-dimensional label system for each adjustable resource of the virtual power plant; dividing the virtual power plant units by adopting a step-by-step division strategy; if the divided virtual power plant units do not pass the capability detection, the virtual power plant units are divided again; and establishing a periodic redivision mechanism, and dynamically adjusting the virtual power plant unit when a redivision condition is triggered. The method solves the problems that traditional virtual power plant unit division depends on fixed geographical boundaries or resource types, is difficult to adapt to power market dynamic requirements and is difficult to respond to power grid local regulation and control in real time, improves power grid interaction precision, enhances market response capability, and realizes dynamic optimization management of the virtual power plant units.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of virtual power plant unit division and adjustable resource aggregation, and particularly relates to a virtual power plant unit dynamic division method and system based on power grid topology. BACKGROUND

[0002] With the accelerated promotion of power market reform and the accelerated construction of new power systems, virtual power plants are actively exploring the normalization of market transactions and grid interaction, and have initially formed a diversified market transaction system covering "medium and long-term transactions + auxiliary service transactions + spot transactions" and demand response.

[0003] At present, in the process of rapid development of virtual power plants, the types of access resources are increasing, and the covered areas are continuously expanding, resulting in insufficient flexibility of virtual power plants as a whole in participating in grid interaction. In cross-regional resource regulation, virtual power plants cannot meet the precise regulation demand of distribution networks. At the same time, in the power market, virtual power plants have always been declared and cleared as a whole, which increases the difficulty of market clearing, limits the ability of virtual power plants to respond to market demand flexibly according to node price differences, and cannot accurately respond to market price signals. SUMMARY

[0004] The purpose of the present application is to provide a virtual power plant unit division method suitable for multi-market collaborative demand and supporting multi-dimensional resource aggregation and dynamic adjustment. Another purpose of the present application is to provide a virtual power plant unit dynamic division method and system based on power grid topology.

[0005] Technical scheme: The virtual power plant unit dynamic division method based on power grid topology comprises the following steps:

[0006] Map the adjustable resources of the virtual power plant to the power grid control nodes to obtain a physical link relationship database and establish a resource-power grid topology mapping model;

[0007] Set a multi-dimensional tag system for each adjustable resource of the virtual power plant, and the multi-dimensional tags include power grid attributes, market attributes and resource characteristics;

[0008] Divide the virtual power plant unit by using a step-by-step division strategy, divide in the first step according to market types, divide in the second step according to power grid topology, and classify in the third step according to resource characteristics;

[0009] Detect the capacity of the divided virtual power plant unit, and when the capacity detection is passed, the virtual power plant unit is formed; when the capacity detection is not passed, the virtual power plant unit is re-divided;

[0010] According to the equipment, resources, market rules and power grid topology structure, a periodic re-division mechanism is established, and when the re-division condition is triggered, the virtual power plant unit is dynamically adjusted.

[0011] Further, a resource-grid topology mapping model is established, specifically as follows:

[0012] According to the user number of the marketing system and the metering point number, the relationship of "transformer (distribution transformer)-house (user, metering point)" is formed; according to the GIS model and the distribution network model data, the relationship of "line (feeder)-transformer (distribution transformer)" is obtained; according to the main distribution model, the virtual power plant resources directly connected to the 110kV bus through the 10kV line and the transformer are aggregated to the 110kV control node, and the other virtual power plant resources connected to the 220kV bus through the 10kV line and the three-winding transformer are aggregated to the 220kV control node, forming the "station-line" relationship, and constructing the "station-line-transformer-house" topology network structure of the virtual power plant adjustable resources.

[0013] Further, the grid attributes include the access voltage level, the control node position and the electrical distance; the market attributes include the market type of the willingness to participate in the market, the declared power supply characteristics and the adjustment capacity; and the resource characteristics include the power generation / load type, the adjustment rate and the duration.

[0014] Further, a step-by-step division strategy is used to divide the virtual power plant units, the first step is to divide according to the market type, to obtain the spot unit, the frequency modulation auxiliary service unit, the peak shaving auxiliary service unit and the demand response unit; the second step is to divide according to the grid topology, the units participating in the adjustment market are subdivided according to the region, and the units participating in the spot market are divided according to the voltage level of the clearing node; and the third step is to classify according to the resource characteristics, and under the same market and topology grouping, the resources are secondarily clustered according to the similarity degree of the adjustment characteristics.

[0015] Further, the virtual power plant units after division are subjected to capacity detection, and the capacity detection items include the adjustment capacity, the adjustment accuracy, the duration and the response time, etc.

[0016] Further, the re-division conditions of the periodic re-division mechanism include the binding / unbinding of important equipment or resources, the major adjustment of market rules and the change of grid topology structure.

[0017] The virtual power plant unit dynamic division system based on the grid topology of the grid topology of the application comprises:

[0018] A topology modeling module is used to construct a resource-grid topology mapping model;

[0019] A resource tag module is used to maintain a multi-dimensional attribute tag database of the virtual power plant adjustable resources, and the multi-dimensional tags include the grid attributes, the market attributes and the resource characteristics;

[0020] A virtual power plant unit management module is used to realize the step-by-step division of the virtual power plant units and support the manual / automatic mode.

[0021] The measurement data acquisition module is used to collect power data and calculate regulation limits on a virtual power plant unit basis.

[0022] Interaction interfaces are used to support standardized data interaction with other external systems.

[0023] Furthermore, in the topology modeling module, a resource-power grid topology mapping model is established, as follows:

[0024] Based on the user ID and metering point ID of the marketing system, a "transformer (distribution transformer) - household (user, metering point)" relationship is formed; based on the GIS model and distribution network model data, a "line (feeder) - transformer (distribution transformer)" relationship is derived; based on the main distribution model, virtual power plant resources directly connected to the 110kV busbar via 10kV lines and transformers are connected to the 110kV control node, and other virtual power plant resources connected to the 220kV busbar via 10kV lines and three-winding transformers are aggregated to the 220kV control node, forming a "station-line" relationship, and constructing a "station-line-transformer-household" topology network structure for the adjustable resources of the virtual power plant.

[0025] Furthermore, in the resource tagging module, the grid attributes include the access voltage level, control node location, and electrical distance; the market attributes include the market type of willingness to participate in the market, the declared power generation and consumption characteristics, and regulation capacity; and the resource characteristics include the power generation / load type, regulation rate, and duration.

[0026] Furthermore, the virtual power plant unit management module, in automatic mode, supports automatic step-by-step partitioning of virtual power plant units. The first step is partitioning by market type, resulting in spot market units, frequency regulation ancillary service units, peak shaving ancillary service units, and demand response units. The second step is partitioning by grid topology, classifying units participating in the regulation market according to their region of origin, and units participating in the spot market according to the clearing node voltage level. The third step is classification by resource characteristics; under the same market and topology grouping, resources are secondary clustered based on the similarity of their external regulation characteristics.

[0027] Beneficial effects: Compared with the prior art, the significant advantages of this invention are: 1. By dividing the grid topology, this invention enables virtual units to accurately respond to local grid control needs, thus improving the accuracy of grid interaction; 2. By supporting flexible market declaration based on node price differences, this invention improves market clearing efficiency and the overall economic benefits of virtual power plants; 3. This invention establishes a periodic re-division mechanism to adapt to changes in resource aggregation and market rules, achieving dynamic optimization management of virtual power plant units; 4. This invention provides a standardized division process and system architecture, facilitating large-scale promotion and application. Attached Figure Description

[0028] Figure 1This is a flowchart of the virtual power plant unit dynamic partitioning method of the present invention;

[0029] Figure 2 A topology diagram of the adjustable resources of a virtual power plant, showing the relationships between stations, lines, transformers, and customers.

[0030] Figure 3 This is a schematic diagram of the structure of a dynamic partitioning system for virtual power plant units. Detailed Implementation

[0031] The invention will now be further described with reference to the accompanying drawings.

[0032] The method for dynamically partitioning virtual power plant units according to the present invention includes the following steps:

[0033] Step S1: By mapping the adjustable resources of the virtual power plant to the grid control nodes, a physical link relationship database is formed, and a resource-grid topology mapping model is established.

[0034] Based on the user ID and metering point ID of the marketing system, a "transformer (distribution transformer) - household (user, metering point)" relationship is formed; based on the GIS model and distribution network model data, a "line (feeder) - transformer (distribution transformer)" relationship is derived; according to the main distribution model, virtual power plant resources directly connected to the 110kV busbar via 10kV lines and transformers are connected to the 110kV control node, while other virtual power plant resources connected to the 220kV busbar via 10kV lines and three-winding transformers are aggregated to the 220kV control node, forming a "station-line" relationship, thus constructing a "station-line-transformer-household" topology network structure for the adjustable resources of virtual power plants. The adjustable resources of virtual power plants are then mapped to the power grid control nodes, forming a physical connection relationship database.

[0035] Step S2: Set up a multi-dimensional tagging system for each adjustable resource in the virtual power plant, including grid attributes, market attributes, and resource characteristics.

[0036] The multi-dimensional labeling system includes grid attributes, such as access voltage level, control node location, and electrical distance; market attributes, such as market type of willingness to participate in the market, declared power generation and consumption characteristics, and regulation capacity; and resource attributes, such as power generation / load type, regulation rate, and duration.

[0037] Step S3: The virtual power plant units are divided using a step-by-step partitioning strategy. The first step is partitioning by market type, the second step is partitioning by grid topology, and the third step is classifying by resource characteristics.

[0038] A step-by-step partitioning strategy is adopted to divide virtual power plant units. The first step is to divide them according to market type, such as spot market units, frequency regulation ancillary service units, peak shaving ancillary service units, and demand response units. The second step is to divide them according to grid topology, further subdividing units participating in the regulation market according to their region and units participating in the spot market according to the clearing node voltage level (220kV / 110kV). The third step is to classify them according to resource characteristics, and under the same market and topology grouping, secondary clustering is performed based on resources with similar external regulation characteristics.

[0039] Step S4: Perform a capability test on the divided virtual power plant units. If the capability test fails, the virtual power plant units are re-divided.

[0040] The capacity testing of the divided virtual power plant units shall refer to the local virtual power plant management regulations and relevant local power market policies, including regulation capacity, regulation accuracy, duration, response time, etc. If the capacity test is passed, a virtual power plant unit is formed. If the capacity test is not passed, the virtual power plant units need to be re-divided according to the step-by-step division strategy.

[0041] Step S5: Establish a periodic re-partitioning mechanism to dynamically adjust the virtual power plant unit when the re-partitioning condition is triggered.

[0042] The repartitioning conditions include binding / unbinding of important equipment or resources, major adjustments to market rules, and changes in the power grid topology. When the repartitioning conditions are triggered, the virtual power plant units are dynamically adjusted.

[0043] The virtual power plant unit dynamic partitioning system of the present invention includes:

[0044] The topology modeling module is used to construct resource-grid topology mapping relationships;

[0045] The resource tagging module is used to maintain a multi-dimensional attribute tag database for the adjustable resources of the virtual power plant;

[0046] The unit management module is used to implement the step-by-step division of virtual power plant units, and supports both manual and automatic modes.

[0047] The measurement data acquisition module is used to collect power data and calculate regulation limits on a virtual power plant unit basis.

[0048] Interaction Interface Module: Supports standardized data interaction with other external systems.

[0049] Furthermore, a "transformer (distribution transformer) - household (user, metering point)" relationship is formed based on the user ID and metering point ID of the marketing system; a "line (feeder) - transformer (distribution transformer)" relationship is derived based on GIS model and distribution network model data; according to the main distribution model, virtual power plant resources directly connected to the 110kV bus via 10kV lines and transformers are aggregated to the 110kV control node, while other virtual power plant resources connected to the 220kV bus via 10kV lines and three-winding transformers are aggregated to the 220kV control node, forming a "station-line" relationship, thus constructing a "station-line-transformer-household" topology network structure for the adjustable resources of virtual power plants. The adjustable resources of virtual power plants are then mapped to the power grid control nodes, forming a physical connection relationship database.

[0050] Furthermore, the system establishes a multi-dimensional tagging system for each adjustable resource in the virtual power plant. This multi-dimensional tagging system includes: grid attributes (such as access voltage level, control node location, and electrical distance); market attributes (such as market type of intended participation, declared power generation and consumption characteristics, and regulation capacity); and resource characteristics (such as generation / load type, regulation rate, and duration). The system supports modifying resource tags.

[0051] Furthermore, the virtual power plant unit management module is used to implement the step-by-step division of virtual power plant units, supporting both manual and automatic modes. The automatic mode supports automatic step-by-step division of virtual power plant units. The first step is division by market type, such as spot market units, frequency regulation ancillary service units, peak shaving ancillary service units, and demand response units. The second step is division by grid topology, further subdividing units participating in the regulation market according to their region, and units participating in the spot market according to the clearing node voltage level (220kV / 110kV). The third step is classification by resource characteristics, performing secondary clustering based on resources with similar external regulation characteristics within the same market and topology group.

[0052] Furthermore, the measurement data acquisition module performs real-time power acquisition and calculates regulation limits on a virtual power plant unit basis; the interaction interface module supports standardized data interaction with other external systems.

Claims

1. A method for dynamically partitioning virtual power plant units based on power grid topology, characterized in that, Includes the following steps: The adjustable resources of the virtual power plant are mapped to the grid control nodes to obtain a physical link relationship database and establish a resource-grid topology mapping model. A multi-dimensional tagging system is set up for each adjustable resource in the virtual power plant. The multi-dimensional tags include grid attributes, market attributes, and resource characteristics. A step-by-step partitioning strategy is adopted to divide virtual power plant units. The first step is to partition them according to market type, the second step is to partition them according to grid topology, and the third step is to classify them according to resource characteristics. The virtual power plant units are then subjected to capacity testing. If the capacity test is passed, a virtual power plant unit is formed; if the capacity test is not passed, the virtual power plant units are re-divided. Based on equipment, resources, market rules, and power grid topology, a periodic re-partitioning mechanism is established to dynamically adjust virtual power plant units when re-partitioning conditions are triggered.

2. The method for dynamically partitioning virtual power plant units based on power grid topology according to claim 1, characterized in that, Establish a resource-power grid topology mapping model, as follows: Based on the user ID and metering point ID of the marketing system, a "transformer (distribution transformer) - household (user, metering point)" relationship is formed; based on the GIS model and distribution network model data, a "line (feeder) - transformer (distribution transformer)" relationship is derived; based on the main distribution model, virtual power plant resources directly connected to the 110kV busbar via 10kV lines and transformers are connected to the 110kV control node, and other virtual power plant resources connected to the 220kV busbar via 10kV lines and three-winding transformers are aggregated to the 220kV control node, forming a "station-line" relationship, and constructing a "station-line-transformer-household" topology network structure for the adjustable resources of the virtual power plant.

3. The method for dynamically partitioning virtual power plant units based on power grid topology according to claim 1, characterized in that, The grid attributes include access voltage level, control node location, and electrical distance; market attributes include market type of willingness to participate in the market, declared power generation and consumption characteristics, and regulation capacity; resource attributes include power generation / load type, regulation rate, and duration.

4. The method for dynamically partitioning virtual power plant units based on power grid topology according to claim 1, characterized in that, A step-by-step partitioning strategy is adopted to divide virtual power plant units. The first step is to divide them by market type, resulting in spot market units, frequency regulation ancillary service units, peak shaving ancillary service units, and demand response units. The second step is to divide them by grid topology, further subdividing units participating in the regulation market according to their region and units participating in the spot market according to the clearing node voltage level. The third step is to classify them by resource characteristics, and under the same market and topology grouping, the resources are secondary clustered according to the similarity of external regulation characteristics.

5. The method for dynamically partitioning virtual power plant units based on power grid topology according to claim 1, characterized in that, The virtual power plant units are then subjected to capacity testing, which includes regulating capacity, regulating accuracy, duration, and response time.

6. The method for dynamically partitioning virtual power plant units based on power grid topology according to claim 1, characterized in that, The re-partitioning conditions for the periodic re-partitioning mechanism include the binding / unbinding of important equipment or resources, major adjustments to market rules, and changes in the power grid topology.

7. A dynamic partitioning system for virtual power plant units based on power grid topology, characterized in that, include: The topology modeling module is used to build resource-grid topology mapping models; The resource tagging module is used to maintain a multi-dimensional attribute tag database for the adjustable resources of the virtual power plant. The multi-dimensional tags include grid attributes, market attributes, and resource characteristics. The virtual power plant unit management module is used to implement the step-by-step division of virtual power plant units and supports manual / automatic modes. The measurement data acquisition module is used to collect power data and calculate regulation limits on a virtual power plant unit basis. Interaction interfaces are used to support standardized data interaction with other external systems.

8. The virtual power plant unit dynamic partitioning system based on power grid topology according to claim 7, characterized in that, In the topology modeling module, a resource-power grid topology mapping model is established, as follows: Based on the user ID and metering point ID of the marketing system, a "transformer (distribution transformer) - household (user, metering point)" relationship is formed; based on the GIS model and distribution network model data, a "line (feeder) - transformer (distribution transformer)" relationship is derived; based on the main distribution model, virtual power plant resources directly connected to the 110kV busbar via 10kV lines and transformers are connected to the 110kV control node, and other virtual power plant resources connected to the 220kV busbar via 10kV lines and three-winding transformers are aggregated to the 220kV control node, forming a "station-line" relationship, and constructing a "station-line-transformer-household" topology network structure for the adjustable resources of the virtual power plant.

9. The virtual power plant unit dynamic partitioning system based on power grid topology according to claim 7, characterized in that, In the resource tagging module, grid attributes include access voltage level, control node location, and electrical distance; market attributes include market type of willingness to participate in the market, declared power generation and consumption characteristics, and regulation capacity; resource characteristics include power generation / load type, regulation rate, and duration.

10. The dynamic partitioning system for virtual power plant units based on power grid topology according to claim 7, characterized in that, The virtual power plant unit management module supports automatic step-by-step partitioning of virtual power plant units in automatic mode. The first step is partitioning by market type, resulting in spot market units, frequency regulation ancillary service units, peak shaving ancillary service units, and demand response units. The second step is partitioning by grid topology, dividing units participating in the regulation market according to their region and units participating in the spot market according to the clearing node voltage level. The third step is classifying by resource characteristics, and under the same market and topology grouping, performing secondary clustering of resources based on the similarity of external regulation characteristics.

Images