A three-dimensional visualization method for a photovoltaic power station and a related device
By employing multi-scale feature extraction and information fusion technologies, the problem of slow traditional 3D modeling has been solved, generating efficient and accurate 3D reality models of photovoltaic power plants, thereby improving operation and maintenance management efficiency and decision-making accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 华能(嘉峪关)新能源有限公司
- Filing Date
- 2024-11-28
- Publication Date
- 2026-05-29
AI Technical Summary
Traditional 3D modeling technology is time-consuming and cannot meet the needs of rapid modeling for large-scale photovoltaic power plants, affecting the efficiency of operation and maintenance management and the accuracy of decision-making.
Image data is processed using multi-scale feature extraction and multi-source image intelligent interpretation algorithms, and point cloud data is processed by combining topological feature extraction and semantic segmentation neural networks. High-confidence annotation information is generated by fusing feature hierarchy and decision hierarchy information, and 3D modeling is performed using a point cloud stitching method based on feature matching.
It enables the accurate and rapid acquisition of information on the location, terrain, and scene of photovoltaic power station equipment, generating realistic and intuitive 3D scene models, thereby improving operation and maintenance efficiency and management level.
Smart Images
Figure CN122115699A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of intelligent management of photovoltaic power plants, and specifically relates to a three-dimensional visualization method and related devices for photovoltaic power plants. Background Technology
[0002] With the transformation of the global energy structure and the severe challenges of climate change, photovoltaic power generation, as a clean and renewable energy source, is being promoted and applied globally at an unprecedented pace. However, as the scale of photovoltaic power plants continues to expand, the complexity of their operation and maintenance management is also increasing dramatically, placing higher demands on operation and maintenance efficiency and management level. Traditional operation and maintenance methods, such as regular manual inspections and simple sensor monitoring, are no longer sufficient to meet the current operational needs of high efficiency and low cost. Manual inspections are not only time-consuming and labor-intensive, but also easily affected by subjective judgment and environmental conditions, making it difficult to achieve comprehensive coverage and accurate assessment; while traditional monitoring systems can usually only provide limited data, making it difficult to support in-depth analysis and prediction.
[0003] To address these issues, 3D modeling and visualization platform technologies have been widely applied in the operation and maintenance management of photovoltaic power plants in recent years. However, traditional 3D modeling techniques, such as manual modeling or modeling methods based on ground measurement data, are not only time-consuming but also struggle to meet the needs of rapid modeling for large-scale photovoltaic power plants. Especially in emergency situations such as the expansion, renovation, or troubleshooting of photovoltaic power plants, the lag in modeling speed often limits the efficiency of operation and maintenance management and the accuracy of decision-making. Summary of the Invention
[0004] The purpose of this invention is to provide a three-dimensional visualization method and related device for photovoltaic power plants, so as to solve the problem of slow three-dimensional modeling speed in the prior art.
[0005] To achieve the above objectives, the present invention adopts the following technical solution: Firstly, a method for three-dimensional visualization of photovoltaic power plants includes the following steps: Collect image data and point cloud data of photovoltaic power plants; After feature extraction from the image data, the equipment location information of the photovoltaic power station is obtained. After feature extraction from the point cloud data, terrain annotation information and scene annotation information are obtained by classification. After hierarchical fusion of the device location information, terrain annotation information and scene annotation information, hierarchical fusion of decision-making is performed to obtain high-confidence annotation information and multi-dimensional device information; A three-dimensional real-scene model of the photovoltaic power station was constructed based on the high-confidence annotation information and multi-dimensional equipment information.
[0006] In some embodiments, the step of extracting features from the image data and then interpreting it to obtain the equipment location information of the photovoltaic power station specifically includes: After processing the image data using multi-scale feature extraction, the equipment location information of the photovoltaic power station is obtained using a multi-source image intelligent interpretation algorithm.
[0007] In some implementations, the step of extracting features from the point cloud data and classifying it to obtain terrain annotation information and scene annotation information specifically includes: After processing the point cloud data using topological feature extraction, a semantic segmentation neural network is used for classification to obtain terrain annotation information and scene annotation information.
[0008] In some implementations, the image data includes orthophoto data and oblique image data.
[0009] In some implementations, the step of performing hierarchical fusion of the device location information, terrain annotation information, and scene annotation information through feature hierarchical fusion to obtain high-confidence annotation information and multi-dimensional device information specifically includes: By fusing the device location information, terrain annotation information, and scene annotation at different scales through feature hierarchy, multi-dimensional device information is obtained. By combining the multi-dimensional device information and fusing the terrain annotation information and scene annotation information through a hierarchical decision fusion method, high-confidence annotation information is obtained.
[0010] In some implementations, the step of constructing a three-dimensional real-scene model of the photovoltaic power station based on the high-confidence annotation information and multi-dimensional device information specifically includes: The key features of the point cloud data are extracted using a point cloud stitching method based on feature matching. Combined with the high-confidence annotation information and multi-dimensional equipment information, a three-dimensional real-scene model of the photovoltaic power station is obtained by three-dimensional modeling.
[0011] Secondly, a three-dimensional visualization system for photovoltaic power plants includes: The data acquisition module is used to collect image data and point cloud data from photovoltaic power plants; The feature extraction module is used to extract features from the image data and then interpret it to obtain the equipment location information of the photovoltaic power station; and to extract features from the point cloud data and then classify it to obtain terrain annotation information and scene annotation information. The data fusion module is used to perform hierarchical fusion of the device location information, terrain annotation information and scene annotation information through feature hierarchical fusion, and then perform decision hierarchical fusion to obtain high-confidence annotation information and multi-dimensional device information. The 3D model building module is used to construct a 3D real-scene model of the photovoltaic power station based on the high-confidence annotation information and multi-dimensional equipment information.
[0012] Thirdly, an electronic device includes a memory, a processor, and a computer program stored in the memory and executable in the processor, wherein the processor executes the computer program to implement the steps of the three-dimensional visualization method for a photovoltaic power station.
[0013] Fourthly, a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the three-dimensional visualization method for a photovoltaic power station.
[0014] Fifthly, a computer program product, the computer product comprising a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the three-dimensional visualization method for a photovoltaic power station.
[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a 3D visualization method for photovoltaic power plants. By processing image data through multi-scale feature extraction and multi-source image intelligent interpretation algorithms, and processing point cloud data using topological feature extraction and semantic segmentation neural networks, it can accurately and quickly obtain equipment location information, terrain annotation information, and scene annotation information of photovoltaic power plants, thereby improving the accuracy and efficiency of 3D visualization.
[0016] Furthermore, this invention utilizes a feature-hierarchical and decision-hierarchical information fusion method to integrate equipment location information, terrain annotation information, and scene annotation information in multiple dimensions, generating high-confidence annotation information and multi-dimensional equipment information, thus providing reliable data support for constructing an accurate and comprehensive 3D real-world model of a photovoltaic power station.
[0017] Furthermore, this invention utilizes a point cloud stitching method based on feature matching to extract key features from point cloud data, and combines high-confidence annotation information and multi-dimensional device information to perform 3D modeling, which can generate realistic and intuitive 3D real-scene models of photovoltaic power plants, helping operation and maintenance personnel to better understand and analyze the operating status of photovoltaic power plants.
[0018] This invention also provides a complete 3D visualization system for photovoltaic power plants, including modules for data acquisition, feature extraction, data fusion, and 3D model construction. This system realizes the systematization and automation of 3D visualization of photovoltaic power plants, reduces the degree of manual intervention, and improves work efficiency. Attached Figure Description
[0019] Figure 1 This is a flowchart illustrating the 3D visualization method for photovoltaic power plants provided in Example 1. Figure 2 This is a schematic diagram of the structure of the photovoltaic power station three-dimensional visualization system provided in Example 2. Detailed Implementation
[0020] To enable those skilled in the art to better understand the present invention, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. The content described herein is for explanation rather than limitation of the present invention.
[0021] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification and claims of this invention are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, systems, products, or devices.
[0022] Example 1 like Figure 1 As shown, a three-dimensional visualization method for photovoltaic power plants includes the following steps: S1 collects image data and point cloud data of the photovoltaic power station, including orthophoto data and oblique image data; The acquired image and point cloud data are preprocessed, including denoising, correction, coordinate transformation, and data cleaning, to ensure data quality and consistency.
[0023] Due to potential interference from environmental noise, equipment errors, and other factors during the acquisition process, image and point cloud data may contain irrelevant or misleading information. Filtering techniques can effectively remove this noise, improving the purity and accuracy of the data.
[0024] Correction includes two aspects: geometric correction and radiometric correction. Geometric correction mainly addresses geometric distortions in images and point cloud data caused by factors such as sensor viewpoint, terrain undulation, and atmospheric refraction, ensuring the accurate location of the data in geospatial space. Radiometric correction focuses on adjusting the brightness and contrast of image data to eliminate color distortions caused by factors such as lighting conditions and differences in shooting time, making the data more visually consistent.
[0025] Coordinate transformation converts image data and point cloud data from their respective acquisition coordinate systems to Global Positioning System (GPS) coordinates, ensuring spatial consistency of the data.
[0026] Data cleaning removes duplicate data from image and point cloud data, fills in missing data, corrects erroneous data, and ensures the integrity and consistency of the dataset.
[0027] S2, after extracting features from the image data, the equipment location information of the photovoltaic power station is obtained by interpretation; after extracting features from the point cloud data, the terrain annotation information and scene annotation information are obtained by classification. A pre-trained Convolutional Neural Network (CNN) is used to extract features from image data at multiple scales. To capture features at different scales in the image, pre-processed image data is input into the CNN at different resolutions or sizes. Then, by constructing a feature pyramid, low-level (high resolution, rich detail) and high-level (low resolution, rich semantics) features are obtained. Since the pre-trained CNN has already learned rich image features on large-scale datasets, these image features are applied to the image data of photovoltaic power plants through transfer learning. By fine-tuning the network parameters, it is better adapted to the specific image features of photovoltaic power plants, thereby improving the accuracy and efficiency of feature extraction. The image feature maps obtained from the multi-scale feature extraction are then input into a multi-source image intelligent interpretation algorithm to analyze the information in the image and identify the equipment and locations of the photovoltaic power plants.
[0028] PointCNN is used to extract topological features from point cloud data. Point cloud data typically consists of a large number of 3D points, each containing coordinate information (x, y, z), color, and intensity. First, the point cloud data is organized into a matrix or graph structure. Then, PointCNN uses a local feature extraction module to capture local geometric features in the point cloud, encoding the local neighborhood features of each point into a fixed-length vector. After extracting the local features, these features are aggregated into a global feature vector to represent the topological structure of the entire point cloud, achieved through a global pooling layer or a global attention mechanism.
[0029] The preprocessed point cloud data is input into a semantic segmentation neural network, and the extracted features are classified using a fully connected layer to obtain the semantic label of each point in the point cloud. Based on the semantic label, the points in the point cloud data are divided into different categories, such as ground, vegetation, buildings, photovoltaic panels, etc., to obtain terrain labeling information and scene labeling information.
[0030] S3, after the device location information, terrain annotation information and scene annotation information are fused in a hierarchical manner by feature hierarchy, a decision-making hierarchical fusion is performed to obtain high-confidence annotation information and multi-dimensional device information; Preprocess the equipment location information, terrain annotation information, and scene annotation information to clean the data, remove noise and redundant information, improve data quality, and ensure that they have a uniform format, coordinate system, and resolution.
[0031] Extract geometric features of the equipment from the equipment location information, such as location coordinates, direction, and height. Extract geometric and topological features of the terrain from the terrain annotation information, such as slope, elevation, and terrain type. Extract semantic features of the scene from the scene annotation information, such as vegetation type, building type, and road type.
[0032] Local features are constructed at different scales, representing the local terrain and scene around the equipment, while global features represent the terrain and scene layout of the entire photovoltaic power station. Multi-resolution analysis methods are used to construct multi-scale features.
[0033] Features from different information sources are fused to generate multi-dimensional device information, which includes the spatial distribution of devices, device type, device status, device performance parameters, etc.
[0034] The terrain and scene annotation information is initially processed using a classifier or regressor to obtain preliminary decision results, including terrain classification labels and scene semantic labels. Then, the preliminary results from different decision models are fused to generate high-confidence annotation information, including accurate terrain classification and accurate scene semantic labels.
[0035] S4. A three-dimensional real-scene model of the photovoltaic power station is constructed based on the high-confidence annotation information and multi-dimensional equipment information.
[0036] A point cloud stitching method based on feature matching is used to extract key features from the point cloud data. These key feature points are typically locations with significant geometric features, such as device edges, terrain inflection points, and building corners. Each detected key feature point is described to generate a feature descriptor, which includes the feature point's location, orientation, scale, and local geometry. Feature descriptors in point cloud data collected from different perspectives or at different times are matched to find identical or similar feature points. Based on the matched feature points, the point cloud data collected from different perspectives or at different times are stitched together to form a complete point cloud dataset.
[0037] High-confidence annotation information is applied to the stitched point cloud dataset to identify different regions and objects within the point cloud data, providing accurate semantic information for 3D modeling. Then, multi-dimensional equipment information is integrated with the point cloud data to construct a 3D model of the equipment, accurately reflecting its actual condition. After these steps, the Delaunay triangulation method is used to convert the point cloud data into a 3D mesh model. Texture information is extracted from the image and mapped onto the 3D mesh model, thereby constructing a 3D reality model of the photovoltaic power station.
[0038] This embodiment utilizes drones to collect visible light images of photovoltaic power stations, generating 3D digital reality models, orthophoto maps, and various vector data with spatial and textural information. This significantly improves efficiency. The 3D data visualization application overturns the traditional 2D base map, using elements such as photovoltaic power station infrastructure, surrounding ecological environment, buildings, and equipment to construct a digital 3D power station as the base map. This enables free access to the 3D world, including stepless zooming, rotation, and dragging, and allows interaction with corresponding business data. By integrating various business data, it forms big data applications. Combined with specific scenarios, it helps users understand on-site data more quickly from macro to micro levels, enabling them to make more timely decisions.
[0039] Example 2 like Figure 2 As shown, this embodiment provides a 3D visualization system for a photovoltaic power station, including: a data acquisition module, a feature extraction module, a data fusion module, and a 3D model construction module; The data acquisition module collects image data and point cloud data from the photovoltaic power station; After the feature extraction module extracts features from the image data, it interprets the equipment location information of the photovoltaic power station. After extracting features from the point cloud data, it classifies the terrain annotation information and scene annotation information. The data fusion module performs hierarchical fusion of the device location information, terrain annotation information and scene annotation information through feature hierarchical fusion to obtain high-confidence annotation information and multi-dimensional device information. The 3D model building module constructs a 3D real-scene model of the photovoltaic power station based on the high-confidence annotation information and multi-dimensional equipment information.
[0040] The 3D visualization system for photovoltaic power plants provided in this embodiment can integrate regular thermal infrared inspection results with real-time data from various sensors to form a truly static and dynamic 3D mirrored world of the power plant. This mirrored world objectively reflects various conditions in the real physical world and, relying on computer processing technology, enables efficient overview, condition monitoring, early warning and forecasting, problem tracking, detailed access, multi-dimensional cross-linking analysis, and data mining. It can also simulate and calculate the actual illumination of light at different times based on the real mirrored world. Future implementations can further support various power plant workflows, automated verification of key nodes, contingency plan simulations, and other multi-dimensional scenarios, steadily advancing power plant management towards full digitalization and intelligence.
[0041] The module division in this embodiment of the invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in the various embodiments of the invention can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.
[0042] This embodiment also provides an electronic device, which includes a processor and a memory. The memory is used to store a computer program (in this embodiment, the computer program includes a computing component and an iterative component, capable of model calculation and model updating). The computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to realize the corresponding method flow or corresponding function. The processor described in this embodiment can be used for the operation of a three-dimensional visualization method for photovoltaic power plants.
[0043] This embodiment also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space, which stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the three-dimensional visualization method for a photovoltaic power station in the above embodiment.
[0044] This embodiment also provides a computer program product, which includes a computer program that, when executed by a processor, implements the corresponding steps of the three-dimensional visualization method for a photovoltaic power station described in the above embodiment.
[0045] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0046] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0047] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0048] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for three-dimensional visualization of photovoltaic power plants, characterized in that, Includes the following steps: Collect image data and point cloud data of photovoltaic power plants; After feature extraction from the image data, the equipment location information of the photovoltaic power station is obtained. After feature extraction from the point cloud data, terrain annotation information and scene annotation information are obtained by classification. After hierarchical fusion of the device location information, terrain annotation information and scene annotation information, hierarchical fusion of decision-making is performed to obtain high-confidence annotation information and multi-dimensional device information; A three-dimensional real-scene model of the photovoltaic power station was constructed based on the high-confidence annotation information and multi-dimensional equipment information.
2. The method for three-dimensional visualization of a photovoltaic power station according to claim 1, characterized in that, The step of extracting features from the image data and then interpreting it to obtain the equipment location information of the photovoltaic power station specifically includes: After processing the image data using multi-scale feature extraction, the equipment location information of the photovoltaic power station is obtained using a multi-source image intelligent interpretation algorithm.
3. The method for three-dimensional visualization of a photovoltaic power station according to claim 1, characterized in that, The step of extracting features from the point cloud data and classifying it to obtain terrain annotation information and scene annotation information specifically includes: After processing the point cloud data using topological feature extraction, a semantic segmentation neural network is used for classification to obtain terrain annotation information and scene annotation information.
4. The method for three-dimensional visualization of a photovoltaic power station according to claim 1, characterized in that, The image data includes orthophoto data and oblique image data.
5. The method for three-dimensional visualization of a photovoltaic power station according to claim 1, characterized in that, The step of performing hierarchical fusion of device location information, terrain annotation information, and scene annotation information through feature hierarchical fusion to obtain high-confidence annotation information and multi-dimensional device information specifically includes: By fusing the device location information, terrain annotation information, and scene annotation information at different scales through feature hierarchy, multi-dimensional device information is obtained. By combining the multi-dimensional device information and fusing the terrain annotation information and scene annotation information through a hierarchical decision fusion method, high-confidence annotation information is obtained.
6. The method for three-dimensional visualization of a photovoltaic power station according to claim 1, characterized in that, The step of constructing a 3D reality model of the photovoltaic power station based on the high-confidence annotation information and multi-dimensional device information specifically includes: The key features of the point cloud data are extracted using a point cloud stitching method based on feature matching. Combined with the high-confidence annotation information and multi-dimensional equipment information, a three-dimensional real-scene model of the photovoltaic power station is obtained by three-dimensional modeling.
7. A three-dimensional visualization system for photovoltaic power plants, characterized in that, include: The data acquisition module is used to collect image data and point cloud data from photovoltaic power plants; The feature extraction module is used to extract features from the image data and then interpret it to obtain the equipment location information of the photovoltaic power station; and to extract features from the point cloud data and then classify it to obtain terrain annotation information and scene annotation information. The data fusion module is used to perform hierarchical fusion of the device location information, terrain annotation information and scene annotation information through feature hierarchical fusion, and then perform decision hierarchical fusion to obtain high-confidence annotation information and multi-dimensional device information. The 3D model building module is used to construct a 3D real-scene model of the photovoltaic power station based on the high-confidence annotation information and multi-dimensional equipment information.
8. An electronic device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable in the processor. When the processor executes the computer program, it implements the steps of the three-dimensional visualization method for a photovoltaic power station as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the three-dimensional visualization method for a photovoltaic power station as described in any one of claims 1 to 6.
10. A computer program product, said computer product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the three-dimensional visualization method for a photovoltaic power station as described in any one of claims 1 to 6.