Power scene multi-source and multi-state model reconstruction method, medium and system

Abstract

The invention discloses a power scene multi-source and multi-state model reconstruction method, medium and system, and the method comprises the following steps: S1, obtaining multi-source heterogeneous data, and constructing a twin according to the multi-source heterogeneous data; s2, obtaining quasi-static data, and constructing an equipment parameterized twin model according to the quasi-static data; and S3, generalizing the parameterized twinborn model, establishing a physical simulation model, obtaining a simulation result according to the physical simulation model, and mapping the simulation result into the twinborn body. The power grid model of the power scene constructed by the model reconstruction method provided by the invention can perform fusion processing on multi-source heterogeneous three-dimensional space data to form a multi-source multi-state power grid three-dimensional digital model, so that a uniform data structure is provided for rapid construction and later operation and inspection of the power grid, and a power grid system platform can conveniently call and analyze.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid model construction, and in particular to a method for reconstructing a multi-source and multi-state model of a power scenario. Background Art

[0002] Different data sources have different modeling methods, and the theories of each modeling method are different, so the modeling results of different data sources are quite different. With the development of digital technology, most large-scale scene modeling now uses multiple data sources for modeling. The limitations of single data source three-dimensional modeling are large. It can still meet the requirements of modeling accuracy in small scenes. In large scenes such as power grid models, there are various types of products and items and complex terrain structures. It is difficult to fully express the model in the scene based on a single data source. Therefore, when constructing large scenes such as power grid models, it is necessary to model based on multiple data sources so that multiple data sources can be integrated into a better twin system. In the twin system, the power grid model can be compatible with multiple model data sources. However, in the twin system, it is easy to generate heterogeneous models due to different formats between different models. The lack of interoperability between the models leads to incomplete loss of data attribute analysis, which makes it difficult to integrate resources in the current multi-source and multi-state digital three-dimensional model structure. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a multi-source and multi-state model reconstruction method for power scenarios to solve the problem of difficulty in integrating modeling resources from multiple data sources.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: a multi-source and multi-state model reconstruction method for power scenarios, comprising the following steps: S1. Acquire multi-source heterogeneous data and construct a twin based on the multi-source heterogeneous data; S2. Acquire quasi-static data, and construct a parameterized twin model of the equipment according to the quasi-static data; S3. Generalize the parameterized twin model, establish a physical simulation model, obtain simulation results according to the physical simulation model, and map the simulation results to the twin.

[0005] Furthermore, in the step S1, multi-source heterogeneous data is obtained, and a twin is constructed according to the multi-source heterogeneous data, which specifically includes the following steps: S11: Collect multi-source heterogeneous data; S12: constructing a neutral data format, and performing standardization processing on the multi-source heterogeneous data according to the neutral data format to obtain standardized data, S13: extracting ground point cloud and performing point cloud preprocessing based on the standardized data to obtain a point cloud, S14: performing point cloud clustering and segmentation on the point cloud to obtain entity objects, performing component-level modeling on the device according to the entity objects to obtain a device model, and describing key components of the device.

[0006] Furthermore, in the step S14, describing the key components of the device specifically includes: describing the geometric characteristics, material data and connection relationships of the device.

[0007] Furthermore, in the step S2, quasi-static data is obtained, and a parametric twin model of the equipment is constructed based on the quasi-static data, specifically including: collecting supplementary information of tiny components, modeling according to the supplementary information of tiny components to obtain a component model; and splicing the component models to form the parametric twin model of the equipment.

[0008] Furthermore, in the step S2, it also includes: simplifying the parameterized twin model.

[0009] Furthermore, simplifying the parameterized twin model includes at least one of the following: Use simplification algorithms to reduce the number of polygons in point clouds or surface models while retaining key geometric features; Extract and summarize the parameters of the component model to form a parametric description; Core features that affect device behavior associated with the parameterized twin model are extracted and retained.

[0010] Furthermore, the step S3 also includes: acquiring real-time data of the device, and calculating the real-time data of the device according to the physical simulation model to obtain a simulation result.

[0011] Furthermore, in the step S3, the simulation result is mapped to the twin, which specifically includes the following steps: S31: Standardizing the format of the simulation results to obtain standardized simulation data; S32: Synchronizing the standardized simulation data with the real-time data based on the timestamp so as to align the standardized simulation data with the real-time data; S33: Mapping the standardized simulation data onto the twin.

[0012] Another object of the present invention is to provide a computer medium, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor; wherein the processor is used to implement a multi-source and multi-state model reconstruction method for a power scenario as described above when executing the computer program.

[0013] Another object of the present invention is to provide a multi-source and multi-state model building system for power scenarios, which is used to implement the model reconstruction method as mentioned above, and the model building system includes a data acquisition module, a twin building module and a simulation module; wherein the data acquisition module is used to collect multi-source heterogeneous data and quasi-static data, the twin building module is used to build a twin based on multi-source heterogeneous data, and the physical twin building module is used to build a simulation model based on quasi-static data, obtain simulation results based on the simulation model, and map the simulation results to the twin.

[0014] The beneficial effect of the present invention is that the model simulation method provided by the present invention constructs a twin based on multi-source heterogeneous data, uses quasi-static data such as geographic information, laser point cloud, oblique photography, etc. as components of the equipment parametric model to construct a parametric twin model, and then establishes a physical simulation model based on the parametric twin model, accesses real-time data, thereby realizing power grid equipment simulation analysis and online high-precision calculation functions, and at the same time combines the calculation results with the twin to realize the in-depth application of fused data.

[0015] The model reconstruction method provided by the present invention is used to construct a power grid model for a power scenario, which can fuse and process multi-source heterogeneous three-dimensional spatial data to form a multi-source and multi-state three-dimensional digital model of a power grid, thereby providing a unified data structure for the rapid construction and subsequent operation and inspection of the power grid, and facilitating the calling and analysis of the power grid system platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flowchart of the steps of the multi-source and multi-state model reconstruction method for power scenarios according to the present invention; Figure 2 This is a structural diagram of the multi-source and multi-state model building system for power scenarios described in the present invention. DETAILED DESCRIPTION

[0017] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0018] Please refer to Figure 1 , a multi-source and multi-state model reconstruction method for power scenarios, comprising the following steps: S1. Acquire multi-source heterogeneous data and construct a twin based on the multi-source heterogeneous data; S2. Acquire quasi-static data, and construct a parameterized twin model according to the quasi-static data; S3. Generalize the parameterized twin model, establish a physical simulation model, obtain simulation results according to the physical simulation model, and map the simulation results to the twin.

[0019] From the above description, it can be seen that the beneficial effect of the present invention is that the model simulation method provided by the present invention constructs a twin based on multi-source heterogeneous data, uses quasi-static data such as geographic information, laser point cloud, oblique photography, etc. as components of the equipment parametric model to construct a parametric twin model, and then establishes a physical simulation model based on the parametric twin model, and accesses real-time data to realize power grid equipment simulation analysis and online high-precision calculation functions, and at the same time combines the calculation results with the twin to realize the in-depth application of fused data.

[0020] The model reconstruction method provided by the present invention is used to construct a power grid model for a power scenario, which can fuse and process multi-source heterogeneous three-dimensional spatial data to form a multi-source and multi-state three-dimensional digital model of a power grid, thereby providing a unified data structure for the rapid construction and subsequent operation and inspection of the power grid, and facilitating the calling and analysis of the power grid system platform.

[0021] Furthermore, in the step S1, multi-source heterogeneous data is obtained, and a twin is constructed according to the multi-source heterogeneous data, which specifically includes the following steps: S11: Collect multi-source heterogeneous data; S12: constructing a neutral data format, and performing standardization processing on the multi-source heterogeneous data according to the neutral data format to obtain standardized data; S13: performing ground point cloud extraction and point cloud preprocessing based on the standardized data to obtain a point cloud; S14: performing point cloud clustering and segmentation on the point cloud to obtain entity objects, performing component-level modeling on the device according to the entity objects to obtain a device model, and describing key components of the device.

[0022] From the above description, it can be seen that the model reconstruction method provided by the present invention can integrate and process multi-source heterogeneous three-dimensional spatial data such as terrain, point cloud, and images, so that their formats, coordinate systems, etc. have a unified and standardized expression form, which can realize the rapid construction of power grid models and provide a unified data interface for later operation and inspection, facilitating the calling and analysis of the power grid system platform.

[0023] Furthermore, in the step S14, describing the key components of the device specifically includes: describing the geometric characteristics, material data and connection relationships of the device.

[0024] Furthermore, in the step S11, collecting multi-source heterogeneous data includes at least one of the following: collecting geographic information data (GIS), collecting equipment spatial location, collecting equipment morphology data, collecting building data, and collecting spatiotemporal data.

[0025] From the above description, it can be seen that the model reconstruction method provided by the present invention can perform global and local multi-level and multi-scale model presentation of the power grid environment. Based on ground point cloud extraction and point cloud preprocessing, it can realize hierarchical reconstruction from single body to whole body, and achieve accurate reproduction of the real scene layout by three-dimensional reconstruction.

[0026] Furthermore, in the step S2, quasi-static data is obtained, and a parameterized twin model is constructed based on the quasi-static data, specifically including: collecting supplementary information of tiny components, modeling according to the supplementary information of tiny components to obtain a component model; and splicing the component models to form the parameterized twin model.

[0027] From the above description, it can be seen that the model reconstruction method provided by the present invention can collect supplementary information of tiny components, perform component-level high-precision modeling on the components in the power grid model, and improve the accuracy and completeness of the power grid model from a fine-grained level.

[0028] Furthermore, in the step S2, it also includes: simplifying the parameterized twin model.

[0029] Furthermore, the parameterized twin model is simplified, including at least one of the following: Use simplification algorithms to reduce the number of polygons in point clouds or surface models while retaining key geometric features; Extract and summarize the parameters of the component model to form a parametric description; Core features that affect device behavior associated with the parameterized twin model are extracted and retained.

[0030] From the above description, it can be seen that the model reconstruction method provided by the present invention can simplify the parameterized twin model and generate an abstract model suitable for physical simulation (such as a finite element model, an electromagnetic field model) to reduce the computational complexity of the physical simulation model.

[0031] Furthermore, the step S3 also includes: acquiring real-time data of the device, and calculating the real-time data of the device according to the physical simulation model to obtain a simulation result.

[0032] From the above description, it can be seen that the physical construction model provided by the present invention can adopt real-time data access and subscription and publishing mechanism for the physical simulation model to realize dynamic adjustment and online simulation analysis functions, so that the power grid model constructed using the model reconstruction method of the present invention can realize full life cycle management.

[0033] Furthermore, in the step S3, the simulation result is mapped to the twin, which specifically includes the following steps: S31: Standardizing the format of the simulation results to obtain standardized simulation data; S32: Synchronizing the standardized simulation data with the real-time data based on the timestamp so as to align the standardized simulation data with the real-time data; S33: Mapping the standardized simulation data onto the twin.

[0034] From the above description, it can be seen that the model reconstruction method provided by the present invention combines the simulation results and the real-time data of the sensors on the power grid model equipment to perform geometric and spatial alignment. Among them, the simulation results can provide an accurate description of the device status, and the real-time data can correct the deviation of the simulation results in the actual operating environment. The combination of the two can improve the authenticity of the twin. At the same time, by inputting real-time external data, a dynamic evaluation of the interaction between the device and the environment can be achieved. The fusion of simulation results and external data in geometry, attributes, and timing can provide support for complex scenario analysis of power grid models.

[0035] Furthermore, in the step S31, it also includes: when the simulation result does not include geographic information, acquiring geographic information data.

[0036] Another object of the present invention is to provide a computer medium, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor; wherein the processor is used to implement a multi-source and multi-state model reconstruction method for power scenarios as described above in the present invention when executing the computer program.

[0037] In addition, if Figure 2 As shown, in order to facilitate the implementation of the multi-source and multi-state model reconstruction method of the above-mentioned power scenario of the present invention, the present invention also discloses a multi-source and multi-state model construction system of the power scenario, and the model construction system can be specifically used to implement the above-mentioned model reconstruction method of the present invention, which can specifically include a data acquisition module, a twin establishment module and a simulation module; wherein the above-mentioned data acquisition module is used to collect multi-source heterogeneous data and quasi-static data, the above-mentioned twin establishment module is used to build a twin according to multi-source heterogeneous data, and the above-mentioned physical twin establishment module is used to establish a simulation model according to the quasi-static data, obtain simulation results based on the simulation model, and map the simulation results to the twin.

[0038] Embodiment 1 Please refer to Figure 1 , Embodiment 1 of the present invention is: to provide a multi-source and multi-state model reconstruction method for power scenarios, which includes step S1: obtaining multi-source heterogeneous data and constructing a twin based on the multi-source heterogeneous data.

[0039] Specifically, the above step S1 includes step S11: collecting multi-source heterogeneous data. The collecting of multi-source heterogeneous data includes at least one of the following: collecting geographic information data (GIS), collecting equipment spatial position, collecting equipment morphology data, collecting building data, and collecting spatiotemporal data.

[0040] Among them, in this embodiment, a scene system entity layer with terrain and surface as the main body can be constructed based on geographic information data; a professional system entity layer with power system equipment and equipment connection interaction channels as the main body can be constructed based on equipment spatial position and equipment form data; a building system entity layer with buildings and building structures as the main body can be constructed based on building data; and a space-time system entity layer with time and weather as the main body can be constructed based on space-time data.

[0041] The method provided in this embodiment can build a twin of power grid equipment and environment based on multi-source heterogeneous data such as meteorological environment, geographic information, equipment structure parameters, and equipment business data.

[0042] In this embodiment, step S1 further includes: Step S12: construct a neutral data format, and perform standardization processing on the multi-source heterogeneous data according to the neutral data format to obtain standardized data.

[0043] Step S13: performing ground point cloud extraction and point cloud preprocessing based on the standardized data to obtain a point cloud; Step S14: clustering and segmenting the point cloud to obtain entity objects, performing component-level modeling on the device according to the entity objects to obtain a device model, and describing the key components of the device. The description of the key components of the device specifically includes: describing the geometric characteristics, material data and connection relationships of the device.

[0044] The method provided in this embodiment can be based on massive three-dimensional data, including three-dimensional digital design results of power grids, three-dimensional models, and multi-source heterogeneous three-dimensional spatial data such as terrain, point clouds, and images for fusion processing, so that the format and coordinate system of multi-source heterogeneous data have a unified and standardized expression form, providing a unified data interface for the rapid construction and subsequent operation and inspection of the power grid, facilitating the call and analysis of the power grid system platform, and enabling the power grid model constructed using this method to display the power grid environment in a global and local multi-level and multi-scale model, realizing hierarchical reconstruction from a single unit to the whole, and achieving accurate reproduction of the real scene layout by three-dimensional reconstruction.

[0045] In this embodiment, the model reconstruction method also includes step S2: acquiring quasi-static data, and constructing a parameterized twin model according to the quasi-static data.

[0046] Specifically, in this embodiment, drones can be used to take aerial photos of equipment accessories and tiny components of the equipment (such as wires, insulators, etc.) in the power grid model to obtain supplementary information of the tiny components, and model the component model based on the supplementary information of the tiny components; the component models are spliced ​​to form a parametric twin model of the power grid equipment.

[0047] The twin construction process of existing technologies focuses on collecting point cloud data and sensor graphics, and lacks the ability to integrate multi-source heterogeneous data such as external information (such as weather, etc.); at the same time, the existing twin construction process lacks the ability to capture the details of small components. This method innovatively uses drone aerial photography and other means to collect information on tiny components, so that this method can support high-precision modeling of tiny components, improve the accuracy and completeness of the model from a fine-grained level, and build an external-oriented high-fidelity model.

[0048] In order to reduce the computational complexity of the power grid model constructed by the method described in this embodiment, in this embodiment, step S2 also includes: simplifying the parameterized twin model. By simplifying the parameterized twin model, an abstract model suitable for physical simulation (such as a finite element model, an electromagnetic field model) can be generated, which can effectively reduce the computational complexity.

[0049] Specifically, the parameterized twin model can be simplified by at least one of the following methods: Use simplification algorithms such as mesh reconstruction and shape extraction to reduce the number of polygons in point clouds or surface models, retaining only key geometric features; Extract and summarize the material properties, connection relationships and other information of the components in the component model to form a parametric description (such as elastic modulus, resistance, thermal conductivity); Extract and retain the core features that affect the behavior of the equipment associated with the parameterized twin model (such as the tension characteristics of the conductor and the wind resistance of the tower).

[0050] In this embodiment, the model reconstruction method also includes step S3: generalizing the parameterized twin model to establish a physical simulation model; acquiring real-time data from device sensors, calculating the real-time data of the device according to the physical simulation model to obtain simulation results, and mapping the simulation results to the twin.

[0051] In this embodiment, the model reconstruction method can adopt real-time data access and subscription publishing mechanism (such as Apache Flink) for the physical simulation model, so that the power grid model constructed according to the method can support dynamic adjustment and online simulation analysis, and realize full life cycle management.

[0052] Existing twin construction relies on offline data to update the model state. The method provided in this embodiment emphasizes real-time data access, and can dynamically distribute and update real-time data streams in real-time databases by introducing a subscription publishing mechanism to achieve high-precision online calculation of the model. The method provided in this embodiment can enhance the adaptability of twins to complex environments and modeling accuracy.

[0053] In step S3 of this embodiment, the simulation result is mapped to the twin, which specifically includes the following steps: S31: Standardizing the format of the simulation results to obtain standardized simulation data; S32: Synchronizing the standardized simulation data with the real-time data based on the timestamp so as to align the standardized simulation data with the real-time data; S33: Mapping the standardized simulation data onto the twin.

[0054] In this embodiment, the simulation results are first standardized in format using methods such as CSV and HDF5; the simulation results include geometric data and attribute data. The standardized simulation data is synchronized with the real-time data from the device sensor based on the timestamp to achieve data alignment. Finally, the attribute data of the simulation results is mapped to the actual device or twin.

[0055] The method provided in this embodiment combines simulation data with external real-time data for geometry and space, and provides an accurate description of the equipment status of the power grid model through simulation calculation of the physical simulation model; wherein, real-time data can correct the deviation between the physical simulation results and the actual operating environment, and combining the simulation results with real-time data can improve the authenticity of the twin. At the same time, the input of real-time data from device sensors can realize the dynamic evaluation of the interaction between the device and the environment, so that the simulation results and real-time data are integrated in geometry, attributes, and timing, providing support for the analysis of complex scenarios.

[0056] In the above step S3, it is also necessary to determine whether the simulation result includes geographic information. If the simulation result does not include geographic information, geographic information data (GIS data) is introduced to supplement the location information of the simulation result.

[0057] The method provided in this embodiment can be used to classify power grid models, establish multi-media heterogeneous models such as overhead transmission lines, cable lines, substation bodies, environments, and ancillary facilities, form a corresponding model library, and use spatial transformation, posture alignment, and comprehensive reconstruction methods to configure and improve the models, forming a comprehensive model building technology and constructing a three-dimensional digital model of a multi-source and multi-state power grid.

[0058] In addition, the method provided in this embodiment can be based on the local update method of the real-scene scanning model of location recognition and repositioning. Through the data feedback of the on-site sensor information source and the alignment of the virtual-real model, it can realize the seamless integration of various real-time perception data such as images / videos with the digital twin, and synchronously update and dynamically grow the digital space twin.

[0059] In summary, applying the model reconstruction method provided by the present invention to construct a power grid model of the power scenario can integrate and process multi-source heterogeneous three-dimensional spatial data to form a multi-source and multi-state three-dimensional digital model of the power grid, providing a unified data structure for the rapid construction and subsequent operation and maintenance of the power grid, and facilitating the call and analysis of the power grid system platform.

[0060] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A multi-source and multi-state model reconstruction method for power scenarios, characterized in that: The following steps are involved: S1. Acquire multi-source heterogeneous data and construct a twin based on the multi-source heterogeneous data; S2. Acquire quasi-static data, and construct a parameterized twin model of the equipment according to the quasi-static data; S3. Generalize the parameterized twin model, establish a physical simulation model, obtain simulation results according to the physical simulation model, and map the simulation results to the twin.

2. The model reconstruction method according to claim 1, characterized in that: In the step S1, multi-source heterogeneous data is obtained, and a twin is constructed according to the multi-source heterogeneous data, which specifically includes the following steps: S11: Collect multi-source heterogeneous data; S12: constructing a neutral data format, and performing standardization processing on the multi-source heterogeneous data according to the neutral data format to obtain standardized data, S13: extracting ground point cloud and performing point cloud preprocessing based on the standardized data to obtain a point cloud, S14: performing point cloud clustering and segmentation on the point cloud to obtain entity objects, performing component-level modeling on the device according to the entity objects to obtain a device model, and describing key components of the device.

3. The model reconstruction method according to claim 2, characterized in that: In the step S14, describing the key components of the device specifically includes: describing the geometric characteristics, material data and connection relationships of the device.

4. The model reconstruction method according to claim 1, characterized in that: In the step S2, quasi-static data is obtained, and a parametric twin model of the equipment is constructed based on the quasi-static data, specifically including: collecting supplementary information of tiny components, modeling according to the supplementary information of tiny components to obtain a component model; splicing the component models to form the parametric twin model of the equipment.

5. The model reconstruction method according to claim 1, characterized in that: In the step S2, it also includes: simplifying the parameterized twin model.

6. The model reconstruction method according to claim 5, characterized in that: Simplifying the parameterized twin model includes at least one of the following: Use simplification algorithms to reduce the number of polygons in point clouds or surface models while retaining key geometric features; Extract and summarize the parameters of the component model to form a parametric description; Core features that affect device behavior associated with the parameterized twin model are extracted and retained.

7. The model reconstruction method according to claim 1, characterized in that: The step S3 also includes: acquiring real-time data of the device, and calculating the real-time data of the device according to the physical simulation model to obtain a simulation result.

8. The model reconstruction method according to claim 1, characterized in that: In the step S3, the simulation result is mapped to the twin, which specifically includes the following steps: S31: Standardizing the format of the simulation results to obtain standardized simulation data; S32: Synchronizing the standardized simulation data with the real-time data based on the timestamp, so that the standardized simulation data is aligned with the real-time data; S33: Mapping the standardized simulation data onto the twin.

9. A computer medium, characterized in that include: A memory, a processor, and a computer program stored in the memory and executable on the processor; wherein the processor is used to implement a multi-source and multi-state model reconstruction method for a power scenario as described in any one of claims 1 to 8 when executing the computer program.

10. A multi-source and multi-state model building system for power scenarios, characterized in that: The model building system is used to implement the model reconstruction method according to any one of claims 1 to 8, and the model building system includes a data acquisition module, a twin building module and a simulation module; wherein the data acquisition module is used to collect multi-source heterogeneous data and quasi-static data, the twin building module is used to build a twin based on multi-source heterogeneous data, and the physical twin building module is used to build a simulation model based on quasi-static data, obtain simulation results based on the simulation model, and map the simulation results to the twin.

Images