Power distribution network topological graph generation method and device based on modular topological data set

Abstract

The invention relates to the technical field of power distribution network planning, and particularly provides a power distribution network topological graph generation method and device based on a modular topological dataset, and the method comprises the steps: carrying out the region division of a target region, and obtaining a power supply region in the target region; taking the area description text information of each power supply area as the input of a pre-trained area classification model to obtain the area type of each power supply area output by the pre-trained area classification model; matching a power distribution network topological graph unit database corresponding to each power supply area based on the area type of each power supply area; and generating a power distribution network topological graph corresponding to each power supply area based on the retrieval index of the power distribution network topological graph unit stored in the power distribution network topological graph unit database corresponding to each power supply area and the retrieval index corresponding to each power supply area. According to the technical scheme provided by the invention, a worker can be assisted in drawing the power distribution network topological graph, and the efficiency of drawing the power distribution network topological graph by the worker is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution network planning, and particularly relates to a power distribution network topology graph generation method and device based on modularized topology dataset. BACKGROUND

[0002] With the transformation of power distribution network to intelligentization and high penetration of new energy environment, the planning and design requirements of power distribution network are increasingly improved. The topology structure of power distribution network is no longer a simple single radial or ring structure, but presents more and more characteristics of multi-loop network interconnection and bidirectional power flow.

[0003] With the development of real city planning, the planning and design of power distribution network often covers multiple power supply areas with different power supply requirements. For example, the planning and design of a power distribution network not only covers the central city, industrial park and important business district, but also covers the town area and even the rural area, so as to provide integrated urban and rural power distribution network planning and design, and further improve the efficiency and rationality of power distribution network planning and design. In this case, when designing the power distribution network topology graph, as the areas covered by the power distribution network are more and more, and the power supply requirements of different areas are not the same, the workload of the planning and design of the power distribution network is more and more, and the work of drawing the power distribution network topology graph is more and more complex and heavy, which leads to the continuous reduction of the efficiency of drawing the power distribution network topology graph. SUMMARY

[0004] In order to overcome the above defects, the present application provides a power distribution network topology graph generation method and device based on modularized topology dataset.

[0005] In a first aspect, a power distribution network topology graph generation method based on modularized topology dataset is provided, which comprises:

[0006] dividing a target area into regions to obtain power supply areas in the target area;

[0007] taking the region description text information of each power supply area as the input of a pre-trained region classification model to obtain the region type of each power supply area output by the pre-trained region classification model;

[0008] matching the power distribution network topology graph unit database corresponding to each power supply area based on the region type of each power supply area;

[0009] generating the power distribution network topology graph corresponding to each power supply area based on the search index of the power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to each power supply area and the search index corresponding to each power supply area.

[0010] Preferably, the training process of the pre-trained region classification model comprises:

[0011] constructing training data by using description text information marked with probabilities of each region type;

[0012] training an initial region classification model by using the training data to obtain the pre-trained region classification model.

[0013] Further, the initial region classification model comprises a word embedding layer, a BERT layer, a full connection layer, and a softmax layer.

[0014] Preferably, the power distribution network topology graph unit comprises modularized power distribution network topology graph unit data, device information of each device in the modularized power distribution network topology graph unit data, and a retrieval index corresponding to the modularized power distribution network topology graph unit data.

[0015] Preferably, the retrieval index comprises at least one of the following: a power supply type, a saturated load density, a region type, a node size, a voltage level, a wiring mode, a power supply mode, new energy configuration information, and energy storage configuration information.

[0016] Further, the saturated load density is as follows:

[0017]

[0018] In the above formula, ρ sat is the saturated load density, L max is the maximum theoretical load per unit area in the power supply region, and A is the area of the power supply region.

[0019] Preferably, the retrieval index of the power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to each power supply region and the retrieval index corresponding to each power supply region are used to generate the power distribution network topology graph corresponding to each power supply region, comprising:

[0020] The power distribution network topology graph unit with the highest similarity between the retrieval index of each power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to the power supply region and the retrieval index corresponding to the power supply region is taken as the power distribution network topology graph corresponding to the power supply region.

[0021] Further, the similarity between the retrieval index of each power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to the power supply region and the retrieval index corresponding to the power supply region comprises:

[0022] The retrieval index of the power distribution network topology graph unit and the retrieval index corresponding to the power supply region are respectively input into a feature extraction network based on a neural network to extract corresponding features i q and i f ;

[0023] The features iq and i f The feature fusion is input to neurons of an output layer through a multi-layer perception, and similarity between the search index of the power distribution network topology graph unit and the search index corresponding to the power supply area is obtained.

[0024] In a second aspect, a power distribution network topology graph generation device based on a modularized topology dataset is provided, and the device comprises:

[0025] A division module is configured to divide a target area into power supply areas.

[0026] An analysis module is configured to input the area description text information of each power supply area into a pre-trained area classification model to obtain the area type of each power supply area output by the pre-trained area classification model.

[0027] A matching module is configured to match a power distribution network topology graph unit database corresponding to each power supply area based on the area type of each power supply area.

[0028] A generation module is configured to generate a power distribution network topology graph corresponding to each power supply area based on the search index of the power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to each power supply area and the search index corresponding to each power supply area.

[0029] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:

[0030] The present application provides a power distribution network topology graph generation method and device based on a modularized topology dataset, which comprises: dividing a target area into power supply areas; inputting the area description text information of each power supply area into a pre-trained area classification model to obtain the area type of each power supply area output by the pre-trained area classification model; matching a power distribution network topology graph unit database corresponding to each power supply area based on the area type of each power supply area; and generating a power distribution network topology graph corresponding to each power supply area based on the search index of the power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to each power supply area and the search index corresponding to each power supply area. The technical solution provided by the present application modularizes the power distribution network topology structure corresponding to power supply areas of different categories by means of modularized design thinking, labels different categories of labeling information, and constructs a modularized structure power distribution network topology graph unit dataset. Thus, the existing power distribution network design and planning scheme information can be more effectively utilized. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1is the main step flow schematic diagram of the power distribution network topology graph generation method based on a modularized topology dataset of an embodiment of the present application.

[0032] Figure 2 is a schematic diagram of a target area Z and a power supply area covered thereby of an embodiment of the present application.

[0033] Figure 3 is a schematic diagram of a region classification model of an embodiment of the present application.

[0034] Figure 4 is a similarity evaluation model schematic diagram of an embodiment of the present application. DETAILED DESCRIPTION

[0035] The specific embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0036] To make the objectives, technical solutions and advantages of embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0037] Embodiment 1

[0038] Reference is made to the accompanying drawings Figure 1 , Figure 1 is the main step flow schematic diagram of the power distribution network topology graph generation method based on a modularized topology dataset of an embodiment of the present application. As shown in Figure 1 , the power distribution network topology graph generation method based on a modularized topology dataset in the embodiments of the present application mainly includes the following steps:

[0039] Step S101: performing region division on a target area to obtain power supply areas in the target area;

[0040] Step S102: taking region description text information of each power supply area as an input of a pre-trained region classification model to obtain a region type of each power supply area output by the pre-trained region classification model;

[0041] Step S103: matching a power distribution network topology graph unit database corresponding to each power supply area based on the region type of each power supply area;

[0042] Step S104: generating a power distribution network topology graph corresponding to each power supply area based on a retrieval index of a power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to each power supply area and a retrieval index corresponding to each power supply area.

[0043] In this embodiment, the power supply area database of the server can be searched according to the region information of the target region, so as to determine the power supply area covered by the target region. As shown in Figure 2 The target region Z corresponds to a plurality of power supply areas, and the region mapping unit searches the power supply area database to retrieve the power supply area information corresponding to the power supply areas Z0-Z5 matched to the target region Z after receiving the target region Z input by the user. 34

[0044] In this embodiment, the training process of the pre-trained region classification model includes:

[0045] The training data is constructed by using the description text information labeled with the probability of each region type.

[0046] The initial region classification model is trained by using the training data to obtain the pre-trained region classification model.

[0047] In one embodiment, the initial region classification model includes a word embedding layer, a BERT layer, a fully connected layer, and a softmax layer.

[0048] In one embodiment, the region types include core urban area, high-tech industrial park, central urban area, economic and technological development zone, urban periphery, developed town area, town, population concentrated rural area, population sparse rural area, remote rural area, and pastoral area.

[0049] In one embodiment, as shown in Figure 3 After the region description text information of the power supply area is preprocessed, such as word segmentation, the segmented words are embedded and then input into BERT to obtain the semantic features corresponding to the region description text information. Then, the semantic features are input into the fully connected layer and then into the softmax classifier, so as to determine the region type corresponding to the power supply area. Wherein h1-h 11 are used to indicate the probabilities corresponding to different region types.

[0050] In this embodiment, the power distribution network topology graph unit includes modularized power distribution network topology graph unit data, device information of each device in the modularized power distribution network topology graph unit data, and a retrieval index corresponding to the modularized power distribution network topology graph unit data.

[0051] In one embodiment, the graph unit data is a modularized power distribution network topology graph, which can display the structure and layout of the power distribution network topology graph unit in an editable graph manner. The graph unit data can be an editable file, so that the designer can adjust and modify the graph unit data in actual work, or insert the graph unit into the power distribution network topology graph being edited during the design of the power distribution network. ​

[0052] The device information records each power distribution device contained in the power distribution network topology module, and records the attribute information of each power distribution device. For example, the device information can be recorded in the form of a list. That is, the device information records each device contained in the power distribution network topology unit and the attribute information of each device in the form of a list. For example, Table 1 below shows an example of a device information list:

[0053] Table 1

[0054] Device ID Device Category Device Attribute Information 001 Substation ... 002 Transformer ... 003 Distribution Line ... ... ... ...

[0055] Referring to Table 1, the first column of the list is the device ID, which uniquely identifies each device in the power distribution network topology unit.

[0056] The second column of the list is the device category, which indicates the device category of each device in the list. In this patent, the device categories of each device in the power distribution network topology unit include substations, transformers, power distribution lines, flexible switches, conventional generating units, new energy generating devices, reactive power compensation devices, energy storage devices, and loads, etc.

[0057] The third column of the list is the device attribute information of the device, and different categories of devices also contain different types of attribute information. For example, Tables 2-10 below exemplarily show the device attribute information of different device categories:

[0058] Table 2 Substation Attributes

[0059]

[0060] Table 3 Transformer Attributes

[0061] Serial Number Attribute Brief Description 1 Belonging Station The name of the substation where the transformer is located 2 Transformer Type Main transformer / distribution transformer 3 Connection Relationship The names of the upper and lower level devices 4 Voltage Grade Primary and secondary side rated voltage 5 Rated Capacity kVA / MVA 6 Voltage Ratio The primary side / secondary side voltage ratio of the transformer 7 Winding Connection Method The connection method of the transformer winding (Y / Y, Δ / Y, Δ / Δ, etc.) 8 Impedance Parameter The value of the internal impedance (resistance and reactance) of the transformer.

[0062] Table 4 Power Distribution Line Attributes

[0063]

[0064] Table 5 Flexible Switch Attributes

[0065] Serial Number Attribute Brief Description 1 Type Circuit breaker / load switch / disconnector 2 Connection Relationship The connection of the switch with the upper / lower level device 3 Rated Capacity The rated working capacity of the switch 4 Voltage Grade The voltage range suitable for the switch.

[0066] Table 6 Conventional Generating Unit Attributes

[0067] Serial Number Attribute Brief Description 1 Access Location / 2 Unit Type Thermal / gas 3 Rated Power / 4 Unit operating cost Describe the economic indicators such as unit operation, maintenance and fuel cost

[0068] Table 7 New Energy Generating Device Attributes

[0069] Serial Number Attribute Brief Description 1 Access Location / 2 Power Generation Type Photovoltaic / wind power, etc. 3 Installed Capacity The rated power generation capacity of the device 4 Rated Power MW 5 Geographical Location The installation site of the device

[0070] Table 8 Reactive Power Compensation Device Attributes

[0071] Serial Number Attribute Brief Description 1 Access Location / 2 Compensation Type Static / dynamic 3 Rated Capacity MVar 4 Access Location /

[0072] Table 9 Energy storage device attributes

[0073]

[0074]

[0075] Table 10 Load attributes

[0076] Serial Number Attribute Brief Description 1 Load Number / 2 Access Location / 3 Load Type Residential / commercial / industrial / electric vehicle 4 Rated Power Consumption Capacity /

[0077] In this embodiment, the search index includes at least one of the following: power supply type, saturated load density, area type, node size, voltage level, wiring mode, power supply mode, new energy configuration information, and energy storage configuration information.

[0078] In one embodiment, 1) the power supply type of the corresponding power supply area includes: A+, A, B, C, D, E, and the like, six levels;

[0079] 2) saturated load density (MW / km2);

[0080] 3) node size;

[0081] 4) voltage level, including: 110kV, 10kV, 35kV, 400V, and the like

[0082] 5) wiring mode, including: petal type, diamond type, closed loop network, ring network wiring, bus section zoning power supply, radial type, open loop wiring, and single feed radial type, and the like;

[0083] 6) power supply mode, including: multi-source interconnection, dual power supply, single power supply;

[0084] 7) new energy configuration: centralized photovoltaic, distributed photovoltaic, small-scale distributed photovoltaic, decentralized rooftop photovoltaic, small amount of household photovoltaic, sporadic distributed photovoltaic, medium-sized wind power, small-sized wind power;

[0085] 8) energy storage configuration: large-scale energy storage system, local energy storage, small amount of decentralized energy storage, local small-scale energy storage, no energy storage configuration.

[0086] In one embodiment, the saturated load density is as follows:

[0087]

[0088] In the above formula, p sat is the saturated load density, L max is the maximum theoretical load per unit area in the power supply area, and A is the area of the power supply area.

[0089] In the embodiment, the power supply area corresponding power distribution network topology graph unit database stores the retrieval index of each power distribution network topology graph unit, and the retrieval index of each power supply area is used to generate the power distribution network topology graph corresponding to each power supply area, including:

[0090] The power supply area corresponding power distribution network topology graph unit database stores the retrieval index of each power distribution network topology graph unit, and the retrieval index of each power supply area is used to generate the power distribution network topology graph corresponding to each power supply area, including:

[0091] In one embodiment, the similarity determination process can be implemented by a neural network-based similarity evaluation module, for example, as shown in Figure 4 ,

[0092] The similarity between the retrieval index of each power distribution network topology graph unit stored in the power supply area corresponding power distribution network topology graph unit database and the retrieval index corresponding to the power supply area is obtained by the process, including:

[0093] The retrieval index Q of the power distribution network topology graph unit and the retrieval index F corresponding to the power supply area are input into the neural network-based feature extraction network to extract the corresponding features i q and i f ;

[0094] The features i q and i f are input into the neurons of the output layer after feature fusion by the multi-layer perception MLP, and the similarity S between the retrieval index of the power distribution network topology graph unit and the retrieval index corresponding to the power supply area output by the neurons of the output layer is obtained.

[0095] Based on the above, the user continues to edit the displayed power distribution network topology on the interface of the client, for example, modifies the devices in each sub-area and the connection relationship between the devices; connect the devices between the sub-areas, etc. Then, after the user completes the editing, the power distribution network topology generation unit generates the power distribution network topology edited by the user. Thus, the drawing work of the power distribution network topology is completed.

[0096] Thus, in this way, during the drawing process of the power distribution network topology by the staff, the present scheme divides the power distribution network topology into multiple sub-areas according to the power supply areas covered by the target area corresponding to the power distribution network, and then retrieves the power distribution network topology unit corresponding to each power supply area from the modular structure power distribution network topology unit database, and embeds the retrieved power distribution network topology unit into the corresponding sub-area. Thus, it assists the staff to draw the power distribution network topology, and improves the efficiency of the staff to draw the power distribution network topology.

[0097] Embodiment 2

[0098] Based on the same inventive concept, the application also provides a power distribution network topology graph generation device based on a modular topology data set, comprising:

[0099] a division module, configured to divide a target region into power supply regions in the target region;

[0100] an analysis module, configured to input the regional description text information of each power supply region into a pre-trained regional classification model to obtain the regional type of each power supply region output by the pre-trained regional classification model;

[0101] a matching module, configured to match the power distribution network topology graph unit database corresponding to each power supply region based on the regional type of each power supply region;

[0102] a generation module, configured to generate the power distribution network topology graph corresponding to each power supply region based on the search index of the power distribution network topology graph unit stored in the power distribution network topology graph unit database corresponding to each power supply region and the search index corresponding to each power supply region.

[0103] Preferably, the training process of the pre-trained regional classification model comprises:

[0104] constructing training data by using the description text information labeled with the probability of each regional type;

[0105] training an initial regional classification model by using the training data to obtain the pre-trained regional classification model.

[0106] Further, the initial regional classification model comprises a word embedding layer, a BERT layer, a full connection layer and a softmax layer.

[0107] Preferably, the power distribution network topology graph unit comprises modular power distribution network topology graph unit data, device information of each device in the modular power distribution network topology graph unit data and a search index corresponding to the modular power distribution network topology graph unit data.

[0108] Preferably, the search index comprises at least one of the following: power supply type, saturated load density, regional type, node size, voltage level, wiring mode, power supply mode, new energy configuration information and energy storage configuration information.

[0109] Further, the saturated load density is as follows:

[0110]

[0111] In the above formula, ρ sat is the saturated load density, L maxThe maximum theoretical load per unit area of the power supply area, A is the area of the power supply area.

[0112] Preferably, the retrieval index of each power supply area corresponding power distribution network topology graph unit stored in the power distribution network topology graph unit database and the retrieval index of each power supply area are used to generate the power distribution network topology graph corresponding to each power supply area, comprising:

[0113] The power supply area corresponding power distribution network topology graph unit database stores the retrieval index of each power distribution network topology graph unit, and the retrieval index of the power supply area corresponding to the power supply area is the highest similarity between the retrieval index of each power supply area corresponding power distribution network topology graph unit stored in the power distribution network topology graph unit database.

[0114] Further, the similarity between the retrieval index of each power supply area corresponding power distribution network topology graph unit stored in the power distribution network topology graph unit database and the retrieval index of the power supply area corresponding to the power supply area is obtained by the process comprising:

[0115] The retrieval index of the power distribution network topology graph unit and the retrieval index of the power supply area corresponding to the power supply area are input into the feature extraction network based on the neural network to extract the corresponding features i q And i f ;

[0116] The features i q And i f are input into the neurons of the output layer through the multi-layer perception after feature fusion, and the similarity between the retrieval index of the power distribution network topology graph unit and the retrieval index of the power supply area corresponding to the power supply area output by the neurons of the output layer is obtained.

[0117] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented 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.

[0118] The present application is described with reference to flowcharts and / or block diagrams according to the method, device (system), and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a machine that implements the flow Figure 1One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0119] 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.

[0120] 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.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for power distribution network topology graph generation based on modular topological dataset, characterized in that, The method comprises: dividing the target area into regions to obtain power supply regions in the target area; inputting the region description text information of each power supply region into a pre-trained region classification model to obtain the region type of each power supply region output by the pre-trained region classification model; matching the power distribution network topology graph unit database corresponding to each power supply region based on the region type of each power supply region; generating the power distribution network topology corresponding to each power supply region based on the search index of the power distribution network topology unit stored in the power distribution network topology graph unit database corresponding to each power supply region and the search index corresponding to each power supply region.

2. The method of claim 1, wherein, The training process of the pre-trained region classification model comprises: constructing training data using description text information labeled with region type probability; training an initial region classification model using the training data to obtain the pre-trained region classification model.

3. The method of claim 2, wherein, The initial region classification model comprises a word embedding layer, a BERT layer, a full connection layer, and a softmax layer.

4. The method of claim 1, wherein, The power distribution network topology unit comprises modular power distribution network topology unit data, device information of each device in the modular power distribution network topology unit data, and a search index corresponding to the modular power distribution network topology unit data.

5. The method of claim 1, wherein, The search index comprises at least one of the following: power supply type, saturated load density, region type, node size, voltage level, wiring method, power supply method, new energy configuration information, and energy storage configuration information.

6. The method of claim 5, wherein, The saturated load density is as follows: In the above formula, p sat is the saturated load density, L max is the maximum theoretical load per unit area in the power supply area, and A is the area of the power supply area.

7. The method of claim 1, wherein, The generation of the power distribution network topology corresponding to each power supply region based on the search index of the power distribution network topology unit stored in the power distribution network topology graph unit database corresponding to each power supply region and the search index corresponding to each power supply region comprises: The power distribution network topology unit corresponding to the power supply region is obtained as the power distribution network topology corresponding to the power supply region by searching the search index of each power distribution network topology unit stored in the power distribution network topology graph unit database corresponding to the power supply region.

8. The method of claim 7, wherein, The similarity between the search index of each power distribution network topology unit stored in the power distribution network topology graph unit database corresponding to the power supply region and the search index corresponding to the power supply region is obtained by: The search index of the power distribution network topology graph unit and the search index corresponding to the power supply area are respectively input to a neural network-based feature extraction network to extract corresponding features i q and i f ; The features i q And i f The feature fusion is input to the neurons of the output layer through the multilayer perception machine, and the similarity between the search indexes of the power distribution network topology graph units and the search indexes corresponding to the power supply area output by the neurons of the output layer is obtained.

9. A device for power distribution network topology map generation based on the modular topological dataset based power distribution network topology map generation method of any of claims 1-8, characterized in that, The device comprises: a division module configured to divide the target area into regions to obtain power supply regions in the target area; an analysis module configured to input the region description text information of each power supply region into a pre-trained region classification model to obtain the region type of each power supply region output by the pre-trained region classification model; a matching module configured to match the power distribution network topology graph unit database corresponding to each power supply region based on the region type of each power supply region; a generation module configured to generate the power distribution network topology corresponding to each power supply region based on the search index of the power distribution network topology unit stored in the power distribution network topology graph unit database corresponding to each power supply region and the search index corresponding to each power supply region.

Images