Substation deployment method and apparatus, electronic device, and computer readable medium

Abstract

Embodiments of the present disclosure disclose a substation deployment method, device, electronic equipment and computer readable medium. A specific embodiment of the method comprises: obtaining a plurality of candidate construction address data and substation construction purposes; generating construction address and device layout information; in response to determining that the ratio of the construction address corresponding to the use area to the substation corresponding to the estimated area is less than the target ratio, for each region, performing the following steps: determining the corresponding sub-layout information; for each device, querying the device data sequence; determining the device data layout; generating a power equipment layout scheme according to the device data layout set; in response to determining that the power equipment layout scheme passes the audit, performing device deployment. The embodiment not only can efficiently and accurately screen out the construction address and device layout information corresponding to the smart substation, but also can realize the precise deployment of the power equipment according to the area of the construction address, so as to guarantee the overall performance and normal operation of the substation.

Description

Technical Field

[0001] Embodiments of this disclosure relate to the field of computer technology, and more specifically to substation deployment methods, apparatus, electronic devices, and computer-readable media. Background Technology

[0002] Currently, with the continuous development of the power industry and the urgent demand for power supply in society, the deployment of substations can effectively solve the problem of power shortage. The site selection of substations and the selection of power equipment are crucial. The common approach to substation deployment is to determine the substation location and various power equipment through expert consultation and setting.

[0003] However, when using the above method, the following technical problems often arise:

[0004] When the area corresponding to the construction address of a smart substation is valid, it is difficult to deploy power equipment through manual settings. This cannot guarantee the overall performance and normal operation of the substation, or the deployed substation may exceed the usable area, resulting in a series of impacts on the local land layout.

[0005] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0007] Some embodiments of this disclosure provide substation deployment methods, apparatuses, electronic devices, and computer-readable media to address one or more of the technical problems mentioned in the background section above.

[0008] In a first aspect, some embodiments of this disclosure provide a substation deployment method, including: acquiring multiple candidate construction address data and substation construction purpose corresponding to a smart substation; generating construction address and equipment layout information based on the multiple candidate construction address data and the substation construction purpose using a role-based collaboration model and a large language model; in response to determining that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, for each region in the region set corresponding to the equipment layout information, performing the generation step: determining the sub-layout information corresponding to the region in the equipment layout information; for each device in the equipment set in the sub-layout information, querying from the equipment database a device data sequence whose device performance matches the device and whose device volume satisfies the device layout corresponding to the device in the sub-layout information; determining the equipment data layout corresponding to the region based on the obtained device data sequence set; generating a power equipment layout scheme for the smart substation under the construction address based on the obtained equipment data layout set; and in response to determining that the power equipment layout scheme passes the review, executing the equipment deployment corresponding to the power equipment layout scheme.

[0009] Secondly, some embodiments of this disclosure provide a substation deployment apparatus, including: an acquisition unit configured to acquire multiple candidate construction address data corresponding to a smart substation and a substation construction purpose; a first generation unit configured to generate construction address and equipment layout information based on the multiple candidate construction address data and the substation construction purpose, using a role-based collaboration model and a large language model; and a first execution unit configured to, in response to determining that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, perform a generation step for each region in the region set corresponding to the equipment layout information: determining the equipment layout. The aforementioned regions in the information correspond to sub-layout information; for each device in the device set of the aforementioned sub-layout information, a device data sequence is queried from the device database that matches the device performance and whose occupied volume satisfies the device layout corresponding to the aforementioned device in the aforementioned sub-layout information; based on the obtained device data sequence set, the device data layout corresponding to the aforementioned region is determined; the second generation unit is configured to generate a power equipment layout scheme for the aforementioned smart substation at the aforementioned construction address based on the obtained device data layout set; the second execution unit is configured to execute the device deployment corresponding to the aforementioned power equipment layout scheme in response to the determination that the power equipment layout scheme has passed the review.

[0010] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, such that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any implementation of the first aspect.

[0011] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0012] The above-described embodiments of this disclosure have the following beneficial effects: The substation deployment method of some embodiments of this disclosure can not only efficiently and accurately filter the construction address and equipment layout information corresponding to a smart substation, but also achieve precise deployment of power equipment based on the area of ​​the construction address, thereby ensuring the overall performance and normal operation of the substation. Specifically, the reason why the overall performance and normal operation of a substation cannot be guaranteed is that, even when the area corresponding to the construction address of the smart substation is valid, it is difficult to deploy power equipment through manual settings, which cannot guarantee the overall performance and normal operation of the substation, or the deployed substation exceeds the usable area, leading to a series of impacts on the local land layout. Based on this, the substation deployment method of some embodiments of this disclosure first obtains multiple candidate construction address data and the construction purpose of the smart substation, so as to facilitate the subsequent selection of the construction address and the generation of equipment layout information. Then, based on the multiple candidate construction address data and the construction purpose of the substation, a role-based collaboration model and a large language model are used to generate construction address and equipment layout information. Here, the role-based collaboration model and the large language model can be used to efficiently and accurately determine the construction address from two aspects. Based on this, a large language model can be used to efficiently generate the equipment layout of the substation corresponding to the construction address. Next, in response to the determination that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, for each region in the set of regions corresponding to the equipment layout information, the following generation steps are performed: First, determine the sub-layout information corresponding to the region in the equipment layout information. Second, for each device in the set of devices in the sub-layout information, query the equipment database for a sequence of device data whose performance matches the device and whose occupied volume meets the corresponding equipment layout requirements in the sub-layout information. Here, given the limited available area corresponding to the construction address, only devices whose performance and occupied volume meet the corresponding equipment layout requirements can be considered as candidate data, thus obtaining the equipment data sequence. Therefore, by considering equipment performance and occupied volume, the performance and area occupancy issues of the smart substation can be effectively solved, ensuring the normal operation and overall performance of the smart substation. Third, based on the obtained set of equipment data sequences, the equipment data layout corresponding to the region can be accurately determined. Furthermore, based on the obtained equipment layout data set, a power equipment layout scheme for the aforementioned smart substation at the aforementioned construction address can be accurately generated. This scheme serves as the basis for subsequent equipment deployment in the smart substation, ensuring its performance and normal operation. Finally, in response to the approval of the power equipment layout scheme, the equipment deployment corresponding to the scheme is executed.In summary, given the limited available area at the construction site of the smart substation, by considering the equipment performance and volume occupied by the equipment, and ensuring that the available area and the performance of the smart substation are not exceeded, a precise power equipment layout scheme can be generated to guarantee the normal operation and overall performance of the subsequently built smart substation. Attached Figure Description

[0013] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0014] Figure 1 This is a flowchart of some embodiments of the substation deployment method according to this disclosure;

[0015] Figure 2 These are schematic diagrams of the structure of some embodiments of the substation deployment apparatus according to this disclosure;

[0016] Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0017] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0018] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0019] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0020] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0021] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0022] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] refer to Figure 1 The diagram illustrates a flow 100 of some embodiments of a substation deployment method according to the present disclosure. The substation deployment method includes the following steps:

[0024] Step 101: Obtain data on multiple candidate construction addresses and the purpose of substation construction for the smart substation.

[0025] In some embodiments, the entity executing the above-described substation deployment method (e.g., electronic equipment) can obtain multiple candidate construction address data corresponding to the smart substation and the substation construction purpose. The smart substation can be a substation equipped with various intelligent functions. In practice, a smart substation can be configured with multiple intelligent agents. Each intelligent agent corresponds to one intelligent function. For example, an intelligent agent can be an agent for equipment fault detection. An intelligent agent can also be an agent for equipment operation and maintenance. The smart substation can also be configured with various highly intelligent automatic inspection robots and maintenance robots. The candidate construction address data can be address data related to the candidate construction address. The candidate construction address can be a single, undetermined address supporting the construction of a smart substation. The address data can include: address area, address elevation information, surrounding configuration, and surrounding landscape. The substation construction purpose can be the purpose of constructing the smart substation. For example, the substation construction purpose can include: load requirements, power supply requirements, and power supply objectives.

[0026] Step 102: Based on the above multiple candidate construction address data and the above substation construction purpose, use a role-based collaboration model and a large language model to generate construction address and equipment layout information.

[0027] In some embodiments, the aforementioned implementing entity can generate construction address and equipment layout information based on the multiple candidate construction address data and the substation construction purpose, utilizing a role-based collaboration model and a large language model. The role-based collaboration model can be an E-CARGO model. The E-CARGO model treats the substation as a complex collaborative system, abstracting functional modules (transformers, energy storage, photovoltaic inverters, intelligent monitoring) as roles in E-CARGO, defining inter-module interaction relationships as groups, and environmental parameters (geography, load, policy) as environments. In practice, through the E-CARGO model and fuzzy comprehensive evaluation methods, a comprehensive score can be achieved for each candidate construction address in various aspects (e.g., economic, technical, and environmental), thus enabling the selection of candidate construction addresses. The large language model can be a large language model pre-trained based on a substation site selection dataset. The construction address can be the address determined for the construction of the intelligent substation. The equipment layout information can be the layout information of various power equipment corresponding to the intelligent substation in the area corresponding to the construction address. The equipment layout information corresponds to a layout that meets the intelligent requirements of the substation. For example, the equipment layout information could be "Area 1 layout: Location 1 layout of power equipment 1, Location 2 layout of power equipment 2, Location 3 layout of power equipment 1, Area 2 layout: Location 2 layout of power equipment 4, Location 5 layout of power equipment 3, Location 7 layout of power equipment 1".

[0028] As an example, firstly, the aforementioned implementing entity can generate a first preliminary construction address sequence based on multiple candidate construction address data and the substation construction purpose through data integration and role modeling, fuzzy evaluation processing of collaborative relationships, and multi-objective optimal site selection (NSGA-II algorithm). Then, it generates prompt information representing the selection of at least one construction address based on the multiple candidate construction address data and the substation construction purpose. Next, the prompt information is input into a large language model to obtain a second preliminary construction address sequence. Then, the construction addresses that overlap with the first and second preliminary construction address sequences and whose sequence positions are earlier are determined. Both the first and second preliminary construction address sequences are ordered in descending order of address priority. Finally, based on the construction addresses, the large language model is used to generate equipment layout information.

[0029] In some optional implementations of certain embodiments, the aforementioned execution entity may generate construction address and equipment layout information based on the aforementioned multiple candidate construction address data and the aforementioned substation construction purpose, utilizing a role-based collaboration model and a large language model, including the following steps:

[0030] The first step is to generate an initial set of construction addresses based on the above candidate address data and the purpose of substation construction, using the above large language model.

[0031] As an example, firstly, address selection prompts are generated to represent the selection of multiple construction addresses based on the aforementioned candidate address data and the substation construction purpose. Then, the address selection prompts are input into the aforementioned large language model to obtain the initial construction address.

[0032] The second step involves using the candidate construction address dataset corresponding to the initial construction address set to filter out construction addresses from the initial construction address set using a role-based collaboration model. Here, the role-based collaboration model can be the model corresponding to the Role-Based Collaboration Model.

[0033] The third step is to generate the equipment layout information based on the above construction address and the corresponding candidate construction address data, using the above large language model.

[0034] As an example, firstly, layout generation prompts are generated for the device layout information corresponding to the aforementioned construction address and candidate construction address data. Then, the layout generation prompts are input into the large language model to obtain the device layout information.

[0035] Step 103: In response to determining that the ratio of the usable area corresponding to the above construction address to the estimated area corresponding to the above smart substation is less than the target ratio, for each area in the above equipment layout information corresponding to the area set, the generation step is executed:

[0036] Step 1031: Determine the sub-layout information corresponding to the above-mentioned area in the above-mentioned equipment layout information.

[0037] In some embodiments, the aforementioned executing entity can determine the sub-layout information corresponding to the aforementioned region in the aforementioned equipment layout information. The estimated area corresponding to the smart substation can be the estimated occupied area. The estimated area corresponding to the smart substation can be set by the substation corresponding to each functional area. The estimated area corresponding to the smart substation can be the estimated occupied area estimated by relevant experts or by a large language model. The target ratio can be a value greater than 1. The target ratio can characterize the availability of the usable area corresponding to the construction address. The larger the value, the more ample the setting of the occupied area corresponding to the smart substation. The equipment layout information can characterize the equipment layout under multiple regions (i.e., a region set). Each region in the region set can be a region divided according to function. The sub-layout information can characterize the layout of power equipment within the region. The total area corresponding to the equipment layout information is the same as the area of ​​each region in the region set.

[0038] As an example, the aforementioned executing entity can determine the aforementioned region-corresponding sub-layout information in the aforementioned device layout information by querying the region-corresponding sub-layout information.

[0039] Step 1032: For each device in the device set in the above sub-layout information, query the device database for a sequence of device data whose performance matches the above device and whose occupied volume satisfies the device layout corresponding to the above device in the above sub-layout information.

[0040] In some embodiments, the aforementioned execution entity can, for each device in the device set of the aforementioned sub-layout information, query from the device database a sequence of device data whose performance matches that device and whose occupied volume satisfies the device layout corresponding to that device in the aforementioned sub-layout information. The device database can be a database storing device data for various types of power equipment. For example, power equipment may include: transformers, high-voltage circuit breakers, surge arresters, and instrument transformers. The device performance of power equipment refers to the actual capabilities exhibited by the equipment during operation, encompassing multiple dimensions such as electrical performance, mechanical performance, and thermal performance, directly affecting the efficiency, reliability, and lifespan of the equipment. In substations, the quality of equipment performance directly relates to the safe and stable operation of the power grid. For example, for a transformer, the corresponding device performance may include: no-load current, short-circuit impedance, and noise level. The occupied volume of the equipment can be the size of the space occupied by the power equipment. Device data is equipment description data related to the power equipment. For example, device data may include: device identifier, device performance value, device origin, device price, and device occupied volume. The device data in the device data sequence can be the device data corresponding to devices whose performance meets preset performance requirements and whose occupied volume satisfies a preset occupied volume. The device data in the device data sequence is sorted from highest to lowest according to the matching degree of the corresponding device. The matching degree can be a value between 0 and 1, with a higher value indicating that the corresponding device is more suitable for deployment.

[0041] In some optional implementations of certain embodiments, the device data in the aforementioned device database includes: device identifier, device performance, and device occupied volume. The device identifier can represent the device's identity information. The device occupied volume can be the physical space volume occupied by the device.

[0042] Optionally, the aforementioned execution entity can query a sequence of device data from the device database that matches the device performance and whose occupied volume satisfies the device layout corresponding to the device in the aforementioned sub-layout information, including the following steps:

[0043] The first step is to filter the equipment layout to determine the equipment performance, equipment volume, and equipment intelligence level conditions for the aforementioned equipment. The equipment volume condition has a higher priority than the equipment performance and intelligence level conditions. The equipment performance condition can be that the equipment performance corresponding to the data is higher than the performance of the equipment corresponding to the aforementioned equipment. The equipment volume condition can be that the equipment volume corresponding to the data does not exceed the performance of the equipment corresponding to the aforementioned equipment. The equipment intelligence level condition can be that the intelligence level of the equipment corresponding to the data is higher than the performance of the equipment corresponding to the aforementioned equipment.

[0044] The second step is to query the aforementioned equipment database for a dataset of equipment that meets the above-mentioned equipment performance and volume requirements. Each piece of equipment in the dataset must have performance no less than the performance of the corresponding device mentioned above, and its volume must not exceed that of the corresponding device.

[0045] The third step is to sort the above equipment dataset according to the equipment performance from high to low, so as to obtain the equipment data sequence.

[0046] Optionally, after querying the device data sequence from the device database that matches the device performance and whose occupied volume satisfies the device layout corresponding to the device in the sub-layout information, the method further includes:

[0047] The first step is to remove the device data that does not meet the above-mentioned device intelligence level conditions from the above device data sequence, and obtain the device data sequence after removal.

[0048] The second step is to determine the device data sequence after the above removal as the device data sequence.

[0049] Step 1034: Determine the device data layout corresponding to the above-mentioned area based on the obtained device data sequence set.

[0050] In some embodiments, the executing entity can determine the device data layout corresponding to the aforementioned region based on the obtained device data sequence set. The device data layout can characterize the arrangement of various power devices within the region.

[0051] As an example, the aforementioned execution entity can directly filter out the highest-performing device data from each device data sequence in the device data sequence set, using it as the target device data to obtain the target device dataset. Then, this target device dataset is used as the device dataset for each device deployed within the region, constructing the device data layout.

[0052] In some optional implementations of certain embodiments, the execution entity can determine the device data layout corresponding to the aforementioned region based on the obtained device data sequence set, including the following steps:

[0053] The first step is to select the device data in the leftmost position from each device data sequence in the above device data sequence set as the target device data.

[0054] The second step involves using the obtained target device dataset as graph labels and adding it to the corresponding regional distribution map to obtain the added distribution map. The graph labels can be added later on the regional distribution map. The added distribution map is a distribution map labeled with the target device data.

[0055] The third step involves adding the set of 3D device models corresponding to the obtained target device dataset to the initial virtual region twin model corresponding to the above-mentioned area, thus obtaining the virtual region twin model. The device 3D model can be the constructed 3D model that represents the external structural information of the target device.

[0056] The fourth step is to bind the above-mentioned distribution map and the above-mentioned virtual region twin model to realize the jump from the distribution map to the twin model.

[0057] Fifth, in response to receiving a click to add a target map label in the distribution map, based on the device identifier corresponding to the target map label, a sub-twin model centered on the target map label in the virtual region twin model corresponding to the device identifier is popped up. The model adjustment of the virtual region twin model is adjusted along with the adjustment of the added distribution map.

[0058] Step 6: Based on the distribution map added above, use the large language model described above to generate reasonable device deployment information for the aforementioned regions. This reasonable device deployment information characterizes whether the deployment of each device within the region is reasonable.

[0059] As an example, firstly, the aforementioned execution entity can generate appropriate prompts to determine whether each device in the distribution map is deployed reasonably. Then, the appropriate prompts are input into a large language model to obtain information on the reasonableness of device deployment.

[0060] Step 7: In response to the determination that the above-mentioned equipment deployment information representation is unreasonable, based on the equipment data subsequences and equipment deployment reasonableness information corresponding to each device in the above-mentioned area, the above-mentioned large language model is used to perform graph adjustment on the above-mentioned added distribution map, resulting in an adjusted distribution map, which serves as the equipment distribution map. The unreasonable representation of equipment deployment reasonableness information can be a situation where the deployment of some devices in the added distribution map is unreasonable. Equipment deployment reasonableness information can include: the reasons for the unreasonable equipment deployment and equipment data. Equipment deployment reasonableness information can also include: the reasons for the unreasonable equipment deployment and the equipment region. The adjusted distribution map can be a distribution map after adjusting the equipment for the unreasonable deployment.

[0061] As an example, firstly, adjustment prompts are generated based on device data subsequences and reasonable device deployment information to adjust the graph. Then, the adjustment prompts are input into the aforementioned large language model to perform graph adjustment using the added distribution map, resulting in the adjusted distribution map, which serves as the device distribution map.

[0062] Step 8: Based on the graph differences between the above-mentioned equipment distribution map and the above-mentioned added distribution map, the above-mentioned virtual region twin model is adaptively adjusted to obtain the adjusted twin model, which serves as the region twin model. Here, the adaptive adjustment can be an adjustment of the model content in the virtual region twin model according to the graph differences.

[0063] The ninth step is to determine the equipment distribution map and the regional twin model as the equipment data layout.

[0064] Step 104: Based on the obtained equipment data layout set, generate the power equipment layout scheme of the above-mentioned intelligent substation at the above-mentioned construction address.

[0065] In some embodiments, the aforementioned executing entity can generate a power equipment layout scheme for the smart substation at the aforementioned construction address based on the obtained equipment data layout set. The power equipment layout scheme at the construction address can be the layout scheme of each power device of the substation located at the construction address.

[0066] As an example, the aforementioned execution entity can directly splice together the various device data layouts in the device data layout set to obtain a power equipment layout scheme.

[0067] In some optional implementations of certain embodiments, the aforementioned execution entity may generate a power equipment layout scheme for the aforementioned smart substation at the aforementioned construction address based on the obtained equipment data layout set, including the following steps:

[0068] The first step is to combine the device distribution maps and regional twin models from each device data layout in the device data layout set to obtain the overall device distribution map and the overall regional twin model. The overall device distribution map can be a distribution map of each pre-deployed device under the overall construction area of ​​the construction address. The overall regional twin model can be an overall twin model of the overall construction area of ​​the construction address.

[0069] As an example, the aforementioned execution entity can combine the various device data layouts in the device data layout set according to regional relationships to create a device distribution map, thus obtaining an overall device distribution map. The twin models of each region are then stitched together to obtain an overall regional twin model.

[0070] The second step is to generate the above power equipment layout scheme based on the above overall equipment distribution map and the above overall regional twin model.

[0071] As an example, the aforementioned implementing entity can perform multimodal integration of the overall equipment distribution map and the aforementioned overall regional twin model to obtain a power equipment layout scheme.

[0072] Step 105: In response to the confirmation that the power equipment layout plan has been approved, the equipment deployment corresponding to the power equipment layout plan is executed.

[0073] In some embodiments, in response to determining that the power equipment layout scheme has passed the review, the aforementioned execution entity may perform the equipment deployment corresponding to the aforementioned power equipment layout scheme.

[0074] In some optional implementations of certain embodiments, the device volume occupied in the device data of the aforementioned device database is determined based on the device's three-dimensional model information, and this three-dimensional model information is determined by an automatic three-dimensional scanning device. The device's three-dimensional model information can be the model information corresponding to the device's three-dimensional structural model. In practice, the device's three-dimensional model information may include: the device's three-dimensional shape, the device's total volume, and the device's outer surface information.

[0075] Here, by considering the overall volume of the equipment, it is possible to fully assess whether the volume of the equipment to be used meets the volume requirements. Considering the external surface information and three-dimensional shape of the equipment allows subsequent quality inspectors to clearly understand the type or model of the equipment to be deployed.

[0076] Optionally, using an automatic 3D scanning device, the 3D model information of the equipment is generated through the following steps:

[0077] The first step involves using the laser scanner included in the aforementioned 3D automatic scanning device to perform precision measurements on the target device, thereby obtaining a point cloud dataset for the target device. This point cloud dataset reflects the 3D structure of the target device.

[0078] The second step is to perform at least one of the following operations on each point cloud data in the above point cloud dataset: data cleaning, noise reduction, and registration, to obtain preprocessed point cloud data.

[0079] The third step is to construct the surface information for the target device. This surface information can be related to the surface structure of the target device. In practice, surface information may include: surface pattern, surface shape, and surface color.

[0080] The fourth step involves generating an initial 3D model of the target device based on the preprocessed point cloud dataset and the device surface information. This initial 3D model provides a preliminary representation of the target device's 3D structure and surface characteristics.

[0081] As an example, the aforementioned execution entity can rely on point cloud processing algorithms (such as ICP registration and Poisson reconstruction) and CAD tools to build a preprocessed point cloud dataset and an initial 3D model corresponding to the aforementioned device surface information.

[0082] The fifth step involves refining the initial 3D model to generate a revised 3D model, which serves as the equipment's 3D model information. This refinement can involve adjusting the details of the initial 3D model.

[0083] Optionally, the above method further includes:

[0084] The first step is to randomly select a number of target devices from the batch target device set during the batch deployment process of the aforementioned target devices as the equipment quality inspection sample set.

[0085] The second step involves using the aforementioned automated 3D scanning device to generate 3D model information for each quality inspection device corresponding to each quality inspection sample in the equipment quality inspection sample set. This batch deployment of target equipment may involve deploying multiple target devices within a substation. The equipment quality inspection sample can be a sample indicating whether the volume of the target equipment meets the standards. The 3D model information of the quality inspection equipment can be the 3D model information determined during the quality inspection process.

[0086] The third step involves determining that the model information error of the 3D model information of each of the aforementioned quality inspection devices is less than a predetermined error, confirming that there is no volume problem in the deployment of the target devices, and continuing to deploy the target devices. The model information error may include volume error. That is, by determining the model information error, the possibility of the target devices exceeding the preset requirements is avoided.

[0087] Optionally, the above-mentioned generation of an initial 3D model of the target device based on the preprocessed point cloud dataset and the device surface information includes the following steps:

[0088] The first step is to construct candidate 3D models based on the preprocessed point cloud dataset mentioned above.

[0089] As an example, the aforementioned execution entity can perform modeling processing on the preprocessed point cloud dataset to obtain candidate 3D models.

[0090] The second step involves controlling a multi-angle camera system to capture multi-angle projection videos of the candidate 3D model projected onto the target location. The multi-angle camera system can be a multi-angle camera processing device. That is, the target location is at the center of the multi-angle camera system. The various cameras corresponding to the multi-angle camera system are deployed around the target location. The candidate 3D model projected onto the target location can be the model structure of the candidate 3D model projected onto the target location. The multi-angle projection video can be video captured from multiple angles during the projection.

[0091] The third step is to control the aforementioned multi-directional camera device to capture multi-directional video of the target device placed at the aforementioned target location.

[0092] The fourth step is to obtain a pre-set set of key orientations. Each key orientation in this set can be predetermined and represents a key part of the target equipment. In practice, different key orientations can be set for different key parts of the target equipment. For example, each key orientation can include: directly opposite, directly behind, directly to the right, and directly to the left.

[0093] The fifth step involves identifying the set of projected video frames from the aforementioned multi-directional projection videos whose corresponding orientation sets correspond to the aforementioned key orientation sets, and also identifying the set of video frames from the aforementioned multi-directional projection videos whose corresponding orientation sets correspond to the aforementioned key orientation sets. There is a one-to-one correspondence between the projected video frames in the projected video frame set and the orientations in the orientation set. The projected video frames can represent the model content of the candidate 3D model under the specified orientation.

[0094] Step 6: Based on the above-mentioned projection video frame set, perform frame extraction processing on the above-mentioned multi-directional projection video to obtain an extracted projection video frame set; and based on the above-mentioned video frame set, perform frame extraction processing on the above-mentioned multi-directional video to obtain an extracted video frame set.

[0095] As an example, the aforementioned execution entity can remove the projected video frame set from the multi-directional projected video to obtain the removed video frame set. Then, according to the azimuth difference at every target degree, the removed video frame set is subjected to frame extraction processing to obtain the extracted projected video frame set. For the generation of the extracted video frame set, please refer to the section on the generation of the extracted projected video frame set.

[0096] Step 7: Combine the above-mentioned projected video frame set and the above-mentioned extracted projected video frame set to generate a combined projected video frame set, and combine the above-mentioned video frame set and the above-mentioned extracted video frame set to generate a combined video frame set.

[0097] Step 8: For each projected video frame in the above combined projection video frame set, perform the following processing steps:

[0098] Sub-step 1: Determine the video frame in the above combined video frame set that is in the same orientation as the above projected video frame, and use it as the target video frame.

[0099] Sub-step 2 involves determining the contour difference information between the projected video frame and the target video frame. This contour difference information characterizes the difference between the model contour in the projected video frame and the device contour in the target video frame.

[0100] As an example, the aforementioned execution entity can extract the projection feature information and device feature information corresponding to the contours of the projected video frame and the target video frame, respectively, through feature extraction. Then, the feature information difference between the projection feature information and the device feature information is determined as the contour gap information.

[0101] In the ninth step, in response to the determination that each contour gap information in the obtained contour gap information set represents no difference, the candidate 3D model and the device surface information are combined accordingly to generate the initial 3D model.

[0102] Optionally, the above-mentioned construction of device surface information for the target device may include the following steps:

[0103] The first step, in response to determining that the target device is a square device, is to determine the surface area corresponding to each face of the target device, thus obtaining a set of surface areas. A face can be any aspect of the target device. A square device has six faces: left, right, front, back, top, and bottom. The surface area can be the area beneath each face. There is a one-to-one correspondence between the surface areas in the set of surface areas and the orientations within each face.

[0104] surface area

[0105] The second step is to perform the following first generation step for each surface area in the above surface area set:

[0106] Sub-step 1: Based on the aforementioned surface area, determine the shooting interval distance corresponding to the device surface. The shooting interval distance can be the distance between the camera device and the target device when shooting on the device surface corresponding to the surface area.

[0107] As an example, the aforementioned executing entity can determine the shooting interval distance for the camera device to shoot, with the aim of the content on the device surface corresponding to 90% of the surface area in the captured image.

[0108] Sub-step 2 involves determining the preset upper and lower shooting distances corresponding to the aforementioned surface area. The preset upper shooting distance can be the maximum shooting distance for photographing the device surface corresponding to the surface area. Using the preset upper shooting distance, global device surface feature information can be extracted for the device surface corresponding to the surface area. The preset lower shooting distance can be the minimum shooting distance for photographing the device surface corresponding to the surface area. More detailed local device surface feature information can be extracted for the device surface corresponding to the surface area.

[0109] Sub-step 3: Determine the shooting orientation of the device surface corresponding to the above surface area, and use it as the target shooting orientation.

[0110] Sub-step 4 instructs the camera device to capture images of the target device at the target shooting angle, within the aforementioned shooting interval distance, the aforementioned preset upper shooting distance, and the aforementioned preset lower shooting distance, thereby obtaining a first captured device surface image, a second captured device surface image, and a third captured device surface image. When shooting at the preset lower shooting distance, the captured images are partial shots of the device surface corresponding to individual surface areas. Therefore, the third captured device surface image is the overall device surface image obtained by combining the various partial shots.

[0111] Sub-step 5 involves performing proportional image segmentation on the first, second, and third camera surface images to obtain the first, second, and third camera surface sub-image sets.

[0112] Sub-step 6: For each first imaging device sub-image in the aforementioned first imaging device sub-image set, perform the following information generation steps:

[0113] The first sub-step involves determining the second and third camera images, which are located at the same segmentation position as the first camera image.

[0114] The second sub-step involves inputting the aforementioned first imaging device surface image into an image feature extraction model to generate image feature information. This image feature extraction model can be a neural network model for extracting image features. In practice, different network structures of image feature extraction models can be used to extract features unique to the surface of the target device, depending on the specific target device. For example, the image feature extraction model could be a multi-layered convolutional layer.

[0115] The third sub-step involves inputting the aforementioned image from the third imaging device into an image detail feature extraction model to generate image detail feature information. This model includes: an image texture detail feature extraction model, an image color feature extraction model, and an image isolated point feature extraction model. The image detail feature extraction model can be a neural network model for extracting image detail feature information. Here, the image detail feature information includes: image texture detail feature information, image color feature information, and image isolated point feature information. The image texture detail feature extraction model is used to extract the texture features of the object's surface in the image, reflecting the grayscale distribution and arrangement patterns of local areas. The image color feature extraction model is used to extract the color distribution and attributes of the image, describing the color information of the object's surface. The image isolated point feature extraction model is used to detect isolated points or abnormal regions in the image, highlighting local salient features. In practice, the image texture detail feature extraction model uses convolutional layers to extract local texture features (such as edges and stripes), pooling layers to compress features and retain important information, and fully connected layers to classify or regress texture categories. The image color feature extraction model first converts RGB to the HSV / Lab color space (which is more in line with human visual perception), and then extracts color features through a CNN. The image isolated point feature extraction model extracts image edges based on convolution operations (such as Canny and Sobel operators), and combines CNN to optimize edge localization accuracy.

[0116] The fourth sub-step involves inputting the aforementioned second-image capture device surface image into the overall image feature information extraction model to generate overall image feature information. This overall image feature information extraction model includes: an image style feature information extraction model and an image pattern feature information extraction model. The image style feature information extraction model can extract the overall style features of the image (such as artistic style, color atmosphere, texture tendency, etc.), reflecting the macroscopic visual style of the image. Based on generative adversarial networks, the image style feature information extraction model can generate images with specific styles while extracting style feature vectors. The image pattern feature information extraction model can extract recurring patterns, structures, or texture patterns (such as stripes, grids, repeating patterns, etc.) in the image, reflecting the local or global regularity of the image. The image pattern feature information extraction model incorporates attention modules (such as SE modules, CBAM modules) to focus on pattern regions and suppress background interference.

[0117] The fifth sub-step involves generating comprehensive image feature information based on the aforementioned image detail feature information, image feature information, and overall image feature information.

[0118] As an example, the aforementioned execution entity can combine image detail feature information, image feature information, and image overall feature information to obtain comprehensive image feature information.

[0119] The sixth sub-step is to generate device surface sub-information based on the above-mentioned comprehensive image feature information.

[0120] As an example, the aforementioned execution entity can generate device surface sub-information based on the aforementioned comprehensive image feature information using a generative model.

[0121] Sub-step 6: Summarize the various equipment surface sub-information in the obtained equipment surface sub-information set to obtain the equipment surface summary information.

[0122] Sub-step 7 adjusts the smoothness of the above-mentioned equipment surface summary information to generate adjusted equipment surface information.

[0123] The third step is to define the obtained adjusted equipment surface information set as the equipment surface information.

[0124] Here, the aforementioned "step one to step three" serves as another inventive point of this disclosure, solving another technical problem: "During the quality inspection of target equipment, because equipment of the same type often has a similar appearance, if the surface information of the equipment is not constructed with sufficient detail, quality inspection errors are prone to occur, leading to insufficient intelligence or low performance of the subsequently deployed equipment." Based on this, when the target equipment is square, this disclosure constructs more detailed surface feature information for each surface by performing multi-dimensional (detailed feature extraction and global feature extraction) on each surface of the target equipment. This makes the subsequently constructed 3D model corresponding to the target equipment more accurate, effectively determining whether the model of the equipment to be inspected matches the model of the 3D model during subsequent equipment quality inspection.

[0125] Optionally, the steps also include:

[0126] The first step is to control the multi-angle camera device to capture multi-angle video of the target device placed at the target location, in response to determining that the target device is not a square device. This video is then used as the target multi-angle video.

[0127] The second step is to perform frame extraction on the above target multi-directional video to obtain a directional video frame sequence.

[0128] The third step is to generate the device frame surface information corresponding to each azimuth video frame in the above azimuth video frame sequence, thus obtaining the device frame surface information sequence. For details on generating the device frame surface information, please refer to the section on generating device surface information.

[0129] The fourth step is to determine the device co-region between every two adjacent azimuth video frames in the above azimuth video frame sequence, and obtain the device co-region information group.

[0130] The fifth step is to fuse the surface information of each device frame in the above device frame surface information sequence according to the obtained device same area information group sequence to generate fused surface information, which is used as the device surface information corresponding to the above target device.

[0131] As an example, firstly, for each adjacent number of device frame surface information subsequences in the device frame surface information sequence, based on the device same-region information group subsequences corresponding to the aforementioned device frame surface information subsequences, a first image generative model is used to generate local device surface information within the shooting azimuth angle range corresponding to the device frame surface information subsequences. Then, the obtained local device surface information is input into a second image generative model to obtain the device surface information. Both the first and second image generative models can be generative and adversarial neural network models.

[0132] Here, the aforementioned "steps one through five" serve as another inventive point of this disclosure, solving another technical problem: "During the quality inspection of target equipment, because equipment of the same type often has a similar appearance, if the surface information of the equipment is not constructed with sufficient precision, quality inspection errors are prone to occur, potentially leading to insufficient intelligence or low performance in subsequently deployed equipment." Based on this, this disclosure, when the target equipment is non-square, determines the surface information of the equipment from various orientations to obtain a subsequence of device frame surface information containing redundant surface information. By determining the information group sequence of the same region of the equipment, redundant surface information in the subsequence of device frame surface information can be removed, thereby achieving accurate construction of the surface information of non-square equipment.

[0133] Optionally, the above-mentioned initial 3D model is refined to generate an adjusted 3D model, including the following steps:

[0134] The first step, in response to determining that there are model connections in the initial 3D model, is to acquire an image of at least one model connection in the initial 3D model. The model connection image can be an image taken from the model connection in the initial 3D model.

[0135] The second step involves performing the following adjustments for each of the at least one model connection image mentioned above:

[0136] Sub-step 1: For the aforementioned target device, capture an image of the target model connection point, where the target model connection point is the model connection point corresponding to the model connection point image. Here, the target model connection point image is an image captured on the target device at the location corresponding to the target model connection point.

[0137] Sub-step 2: For the image at the model connection point, perform the following second generation step:

[0138] The first sub-step involves inputting the images at the connection points of the aforementioned models and the target model into the image authenticity discrimination model to generate a discrimination result. The image authenticity discrimination model can be trained adversarially between the generator and the discriminator to learn the feature distribution of forged images, which is then used to detect fake content generated by GANs. In practice, the image authenticity discrimination model can extract general features of forged images through progressive generative adversarial networks, enabling it to detect fake images generated by various GANs across different models. The image authenticity discrimination model can be a ProGAN model.

[0139] The second sub-step, in response to determining that the above discrimination result characterizes the image at the model connection point as an image of the printed model, uses an image generation model to adjust the image at the model connection point based on the image difference information included in the discrimination result, obtaining the adjusted image. The discrimination result also includes: image difference information between the target model connection point image and the image at the model connection point. The image generation model can be a generative model that adjusts the image details of the model connection point image based on the image difference information. In practice, the image generation model can include: an image difference information vector conversion layer, a model connection point image extraction layer, an image difference information vector and model connection point vector fusion layer, and an encoding and decoding layer. The encoding and decoding layer can be a network layer that regenerates the image by encoding the fused vector and then decoding it based on the encoded vector using an attention mechanism.

[0140] As an example, the aforementioned execution entity can first input image difference information and images at model connection points into the image generation model to obtain the adjusted image.

[0141] The third sub-step is to store the image of the model connection point in response to determining that the above discrimination result indicates that the image of the target model connection point is the image of the printed model.

[0142] The fourth sub-step involves using the adjusted image as the image at the model connection point and continuing with the second generation step.

[0143] Sub-step 3: Determine the image at the connection point of the stored model as the image to be replaced.

[0144] The third step is to adjust the connection points of the initial 3D model based on the obtained set of images to be replaced, and obtain the adjusted 3D model.

[0145] As an example, the aforementioned execution entity can, based on the set of images to be replaced, utilize a first image generative model and a second image generative model to regenerate the regenerated device surface information corresponding to the initial 3D model. Then, based on the regenerated device surface information, it performs surface information replacement on the initial 3D model to generate the adjusted 3D model.

[0146] Here, the aforementioned "step one to step three" serves as one of the inventive points of this disclosure, solving another technical problem: "the problem of inaccurate surface information construction at model connection points." Based on this, through an image authenticity discrimination model and an image generation model, the accuracy of surface information at model connection points can be determined, and based on the determination results, the surface information at model connection points can be reconstructed. Thus, the initial 3D model can be refined for each model connection point.

[0147] The above-described embodiments of this disclosure have the following beneficial effects: The substation deployment method of some embodiments of this disclosure can not only efficiently and accurately filter the construction address and equipment layout information corresponding to a smart substation, but also achieve precise deployment of power equipment based on the area of ​​the construction address, thereby ensuring the overall performance and normal operation of the substation. Specifically, the reason why the overall performance and normal operation of a substation cannot be guaranteed is that, even when the area corresponding to the construction address of the smart substation is valid, it is difficult to deploy power equipment through manual settings, which cannot guarantee the overall performance and normal operation of the substation, or the deployed substation exceeds the usable area, leading to a series of impacts on the local land layout. Based on this, the substation deployment method of some embodiments of this disclosure first obtains multiple candidate construction address data and the construction purpose of the smart substation, so as to facilitate the subsequent selection of the construction address and the generation of equipment layout information. Then, based on the multiple candidate construction address data and the construction purpose of the substation, a role-based collaboration model and a large language model are used to generate construction address and equipment layout information. Here, the role-based collaboration model and the large language model can be used to efficiently and accurately determine the construction address from two aspects. Based on this, a large language model can be used to efficiently generate the equipment layout of the substation corresponding to the construction address. Next, in response to the determination that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, for each region in the set of regions corresponding to the equipment layout information, the following generation steps are performed: First, determine the sub-layout information corresponding to the region in the equipment layout information. Second, for each device in the set of devices in the sub-layout information, query the equipment database for a sequence of device data whose performance matches the device and whose occupied volume meets the corresponding equipment layout requirements in the sub-layout information. Here, given the limited available area corresponding to the construction address, only devices whose performance and occupied volume meet the corresponding equipment layout requirements can be considered as candidate data, thus obtaining the equipment data sequence. Therefore, by considering equipment performance and occupied volume, the performance and area occupancy issues of the smart substation can be effectively solved, ensuring the normal operation and overall performance of the smart substation. Third, based on the obtained set of equipment data sequences, the equipment data layout corresponding to the region can be accurately determined. Furthermore, based on the obtained equipment layout data set, a power equipment layout scheme for the aforementioned smart substation at the aforementioned construction address can be accurately generated. This scheme serves as the basis for subsequent equipment deployment in the smart substation, ensuring its performance and normal operation. Finally, in response to the approval of the power equipment layout scheme, the equipment deployment corresponding to the scheme is executed.In summary, given the limited available area at the construction site of the smart substation, by considering the equipment performance and volume occupied by the equipment, and ensuring that the available area and the performance of the smart substation are not exceeded, a precise power equipment layout scheme can be generated to guarantee the normal operation and overall performance of the subsequently built smart substation.

[0148] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a substation deployment device, which are similar to... Figure 1 Corresponding to the method embodiments shown, this substation deployment device can be specifically applied to various electronic devices.

[0149] like Figure 2 As shown, a substation deployment device 200 includes: an acquisition unit 201, a first generation unit 202, a first execution unit 203, a second generation unit 204, and a second execution unit 205. The acquisition unit 201 is configured to acquire multiple candidate construction address data corresponding to a smart substation and the substation construction purpose; the first generation unit 202 is configured to generate construction address and equipment layout information based on the multiple candidate construction address data and the substation construction purpose, using a role-based collaboration model and a large language model; the first execution unit 203 is configured to, in response to determining that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, perform a generation step for each region in the region set corresponding to the equipment layout information: determining the sub-layout corresponding to the region in the equipment layout information. Bureau information; for each device in the device set in the above sub-layout information, query the device database for device data sequences whose performance matches the above device and whose occupied volume meets the device layout corresponding to the above device in the above sub-layout information; determine the device data layout corresponding to the above area based on the obtained device data sequence set; the second generation unit 204 is configured to generate the power equipment layout scheme of the above smart substation at the above construction address based on the obtained device data layout set; the second execution unit 205 is configured to execute the device deployment corresponding to the above power equipment layout scheme in response to the determination that the power equipment layout scheme has passed the review.

[0150] It is understandable that the units described in the substation deployment device 200 are consistent with the reference. Figure 1 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the substation deployment device 200 and the units contained therein, and will not be repeated here.

[0151] The following is for reference. Figure 3It shows a schematic diagram of the structure of an electronic device (e.g., an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0152] like Figure 3 As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0153] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0154] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0155] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0156] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0157] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: acquire multiple candidate construction address data and the construction purpose of the smart substation; generate construction address and equipment layout information based on the multiple candidate construction address data and the construction purpose of the substation, using a role-based collaboration model and a large language model; in response to determining that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, for each region in the region set corresponding to the equipment layout information, perform the following generation steps: determine the sub-layout information corresponding to the region in the equipment layout information; for each device in the equipment set in the sub-layout information, query the equipment database for a sequence of equipment data whose performance matches the device and whose occupied volume satisfies the equipment layout corresponding to the device in the sub-layout information; determine the equipment data layout corresponding to the region based on the obtained set of equipment data sequences; generate a power equipment layout scheme for the smart substation under the construction address based on the obtained set of equipment data layouts; and in response to determining that the power equipment layout scheme passes the review, execute the equipment deployment corresponding to the power equipment layout scheme.

[0158] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0159] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0160] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor can be described as including an acquisition unit, a first generation unit, a first execution unit, a second generation unit, and a second execution unit. The names of these units do not necessarily limit the specific unit; for example, the acquisition unit can also be described as "a unit that acquires data on multiple candidate construction addresses corresponding to a smart substation and the purpose of substation construction."

[0161] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0162] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A substation deployment method, comprising: Obtain data on multiple candidate construction sites for smart substations and the purpose of substation construction; Based on the multiple candidate construction address data and the substation construction purpose, construction address and equipment layout information are generated using a role-based collaboration model and a large language model. In response to the determination that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, for each area in the area set corresponding to the equipment layout information, the following generation steps are performed: Determine the sub-layout information corresponding to the region in the device layout information; For each device in the device set in the sub-layout information, query the device database for a sequence of device data that matches the device performance and whose occupied volume satisfies the device layout corresponding to the device in the sub-layout information; Based on the obtained device data sequence set, the device data layout corresponding to the region is determined. This determination includes: for each device data sequence in the device data sequence set, selecting the leftmost device data as the target device data; adding the obtained target device dataset as a map label to the region distribution map corresponding to the region, resulting in an added distribution map; adding the device 3D model set corresponding to the obtained target device dataset to the initial virtual region twin model corresponding to the region, resulting in a virtual region twin model; binding the added distribution map and the virtual region twin model to achieve a jump from the distribution map to the twin model; and responding to a click on the target map label in the added distribution map, determining the device label corresponding to the target map label... The system identifies and pops up a sub-twin model centered on the target graph label in the virtual region twin model corresponding to the device identifier. The model adjustment of the virtual region twin model is adjusted along with the adjustment of the added distribution map. Based on the added distribution map, the system uses the large language model to generate reasonable device deployment information corresponding to the region. In response to the determination that the reasonable device deployment information representation is unreasonable, the system uses the large language model to adjust the added distribution map based on the device data sub-sequences and reasonable device deployment information corresponding to each device in the region, obtaining an adjusted distribution map as the device distribution map. Based on the graph differences between the device distribution map and the added distribution map, the system adaptively adjusts the virtual region twin model to obtain an adjusted twin model as the region twin model. The system determines the device distribution map and the region twin model as the device data layout. Based on the obtained equipment data layout set, a power equipment layout scheme for the intelligent substation at the construction address is generated; In response to the approval of the power equipment layout plan, the equipment deployment corresponding to the power equipment layout plan is executed.

2. The method according to claim 1, wherein, The step of generating construction address and equipment layout information based on the multiple candidate construction address data and the substation construction purpose, using a role-based collaboration model and a large language model, includes: Based on the multiple candidate construction address data and the substation construction purpose, an initial construction address set is generated using the large language model; Based on the candidate construction address dataset corresponding to the initial construction address set, a role-based collaboration model is used to filter construction addresses from the initial construction address set; Based on the construction address and the corresponding candidate construction address data, the equipment layout information is generated using the large language model.

3. The method according to claim 1, wherein, The device data in the device database includes: device identifier, device performance, device occupied volume; and The step of querying the device data sequence from the device database that matches the device performance and whose occupied volume satisfies the device layout corresponding to the device in the sub-layout information includes: The device performance conditions, device volume conditions, and device intelligence level conditions corresponding to the device are selected from the corresponding device layout. Among them, the device volume condition has a higher priority than the device performance condition and the device intelligence level condition. Query the device database to find a dataset of devices that meet the device performance conditions and the device volume requirements; The device dataset is sorted according to device performance from highest to lowest to obtain a device data sequence; and After querying the device data sequence from the device database that matches the device performance and whose occupied volume satisfies the device layout corresponding to the device in the sub-layout information, the method further includes: Remove the device data that does not meet the device intelligence level condition from the device data sequence to obtain the device data sequence after removal; The removed device data sequence is determined as the device data sequence.

4. The method according to claim 3, wherein, The step of generating a power equipment layout scheme for the smart substation at the construction address based on the obtained equipment data layout set includes: The device distribution map and regional twin model in each device data layout of the device data layout set are combined to obtain the overall device distribution map and the overall regional twin model. The power equipment layout scheme is generated based on the overall equipment distribution map and the overall regional twin model.

5. The method according to claim 1, wherein, The equipment volume occupied in the equipment database is determined based on the equipment's 3D model information, which is determined by an automatic 3D scanning device. The 3D model information is generated using this automatic 3D scanning device through the following steps: The laser scanner included in the device's 3D automatic scanning apparatus is used to perform precision measurements on the target device in order to obtain a point cloud dataset for the target device. Perform at least one of the following operations—data cleaning, noise reduction, and registration—on each point cloud data in the point cloud dataset to obtain a preprocessed point cloud dataset; Construct device surface information for the target device; Based on the preprocessed point cloud dataset and the device surface information, an initial 3D model for the target device is generated. The initial 3D model is refined to generate an adjusted 3D model, which serves as the 3D model information of the device. The method also includes: In response to the batch deployment of the target devices, a number of target devices are randomly selected from the batch target device set as a device quality inspection sample set. Using the aforementioned equipment 3D automatic scanning device, 3D model information of each quality inspection equipment corresponding to each equipment quality inspection sample in the equipment quality inspection sample set is generated; In response to determining that the model information error of the three-dimensional model information of each quality inspection device is less than a predetermined error, it is determined that there is no volume problem in the deployment of the target device, and the deployment of the target device continues.

6. The method according to claim 5, wherein, The step of generating an initial 3D model for the target device based on the preprocessed point cloud dataset and the device surface information includes: Based on the preprocessed point cloud dataset, construct candidate 3D models; Control the multi-angle camera device to capture multi-angle projection video of the candidate 3D model projected onto the target location; Control the multi-angle camera device to capture multi-angle video of the target device placed at the target location; Obtain a pre-set set of key locations; The set of projected video frames in the multi-directional projection video whose corresponding orientation set is the key orientation set is determined, and the set of video frames in the multi-directional video whose corresponding orientation set is the key orientation set is determined. Based on the projected video frame set, the multi-directional projected video is subjected to frame extraction processing to obtain an extracted projected video frame set; and based on the video frame set, the multi-directional video is subjected to frame extraction processing to obtain an extracted video frame set. The projected video frame set and the extracted projected video frame set are combined to generate a combined projected video frame set, and the video frame set and the extracted video frame set are combined to generate a combined video frame set. For each projected video frame in the combined projected video frame set, the following processing steps are performed: The video frame in the combined video frame set that is in the same orientation as the projected video frame is identified as the target video frame. Determine the contour difference information between the projected video frame and the target video frame; In response to the determination that each contour gap information in the obtained contour gap information set represents no difference, the candidate 3D model and the device surface information are combined accordingly to generate the initial 3D model.

7. A substation deployment device, comprising: The acquisition unit is configured to acquire data on multiple candidate construction addresses and the purpose of substation construction for smart substations. The first generation unit is configured to generate construction address and equipment layout information based on the multiple candidate construction address data and the substation construction purpose, using a role-based collaboration model and a large language model. The first execution unit is configured to, in response to determining that the ratio of the usable area corresponding to the construction address to the estimated area corresponding to the smart substation is less than a target ratio, perform the generation step for each region in the region set corresponding to the equipment layout information: determining the sub-layout information corresponding to the region in the equipment layout information; For each device in the device set of the sub-layout information, query the device database for a device data sequence whose performance matches the device and whose occupied volume satisfies the device layout corresponding to the device in the sub-layout information; determine the device data layout corresponding to the region based on the obtained device data sequence set, wherein determining the device data layout corresponding to the region based on the obtained device data sequence set includes: for each device data sequence in the device data sequence set, selecting the device data in the leftmost position from the device data sequence as the target device data; adding the obtained target device dataset as a map label to the regional distribution map corresponding to the region to obtain the added distribution map; adding the device 3D model set corresponding to the obtained target device dataset to the initial virtual region twin model corresponding to the region to obtain the virtual region twin model; binding the added distribution map and the virtual region twin model to achieve the jump from the distribution map to the twin model; responding to the click of the target map label in the added distribution map, determining the device label corresponding to the target map label based on the device label in the added distribution map. The system identifies and pops up a sub-twin model centered on the target graph label in the virtual region twin model corresponding to the device identifier. The model adjustment of the virtual region twin model is adjusted along with the adjustment of the added distribution map. Based on the added distribution map, the system uses the large language model to generate reasonable device deployment information corresponding to the region. In response to the determination that the reasonable device deployment information representation is unreasonable, the system uses the large language model to adjust the added distribution map based on the device data sub-sequences and reasonable device deployment information corresponding to each device in the region, obtaining an adjusted distribution map as the device distribution map. Based on the graph differences between the device distribution map and the added distribution map, the system adaptively adjusts the virtual region twin model to obtain an adjusted twin model as the region twin model. The system determines the device distribution map and the region twin model as the device data layout. The second generation unit is configured to generate a power equipment layout scheme for the smart substation at the construction address based on the obtained equipment data layout set. The second execution unit is configured to execute the equipment deployment corresponding to the power equipment layout scheme in response to the determination that the power equipment layout scheme has passed the review.

8. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-6.

9. A computer-readable medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

Images