Intelligent generation method and system for wind power plant inspection path based on digital terrain model

Abstract

The invention relates to the technical field of new energy wind power generation, in particular to a digital terrain model-based wind power plant inspection path intelligent generation method and system, and the method constructs a digital terrain model and divides terrain grid units by integrating multi-source terrain data such as unmanned aerial vehicle laser radar scanning, satellite remote sensing images and ground surveying and mapping. A multi-layer cooperative feasible region of a ground passage layer and an air flight layer is established based on inspection equipment characteristics, a communication blind area is identified by combining a hierarchical communication reachability model, and a comprehensive cost field containing cooperative cost is constructed by fusing terrain passage and communication characteristics. A double-layer collaborative optimization algorithm is adopted, a task distribution sequence is generated under the constraint of task priority and a time window, an optimal inspection path is formed under the constraint of a comprehensive cost field through path planning and smooth correction, and efficient, safe and intelligent inspection of the wind power plant is achieved.

Description

A Method and System for Intelligent Generation of Wind Farm Inspection Paths Based on Digital Terrain Models Technical Field

[0001] This invention relates to the field of new energy wind power generation technology, specifically to a method and system for intelligent generation of wind farm inspection paths based on digital terrain models. Background Technology

[0002] Wind farm inspection is a crucial step in ensuring the safe, stable, and efficient operation of wind power equipment. Its main tasks include equipment visual inspection, component condition assessment, environmental safety monitoring, and operational data collection. With the rapid expansion of wind power capacity, wind farms are increasingly located in mountainous areas, canyons, plateaus, and offshore regions, facing complex and variable terrain conditions and highly volatile meteorological environments, posing significant challenges to the planning and execution of inspection tasks. Traditional manual inspections and single-drone autonomous flight methods have significant shortcomings in terms of operational safety, coverage integrity, and communication stability, making them unsuitable for the intelligent operation and maintenance needs of large-scale wind farms.

[0003] Currently, path planning for wind farm inspection tasks both domestically and internationally primarily employs ground path generation based on Geographic Information Systems (GIS) or aerial path planning based on flight path algorithms. Common methods include A* algorithm, Dijkstra's algorithm, genetic algorithm, ant colony algorithm, and terrain feasible region analysis based on digital elevation models. These methods typically rely on static two-dimensional maps or single-dimensional constraints, failing to fully consider complex factors such as terrain undulations, wind corridor effects, and communication blind spots. This leads to problems in practical applications, including insufficient accessibility, discontinuous data links, and delayed risk assessment. Furthermore, existing systems often plan UAV and ground vehicle inspections separately, lacking collaborative scheduling and dynamic feedback mechanisms, resulting in low overall efficiency and insufficient risk control, failing to meet the needs of intelligent inspection of the entire wind farm area.

[0004] Therefore, there is an urgent need for a method and system for intelligent generation of wind farm inspection paths that combines digital terrain models. This system should be able to comprehensively consider terrain features, communication conditions, and wind corridor risks, and achieve multi-source information fusion and dynamic path optimization to improve the safety, continuity, and intelligence of wind farm inspection tasks. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for intelligent generation of wind farm inspection paths based on digital terrain models, which addresses the shortcomings of the prior art. This method solves the technical problem that existing systems often plan inspections of UAVs and ground vehicles separately, lacking collaborative scheduling and dynamic feedback mechanisms, resulting in low overall efficiency and insufficient risk control.

[0006] The objective of this invention is achieved through the following technical solution: Firstly, this invention provides an intelligent method for generating wind farm inspection paths based on a digital terrain model, comprising: acquiring multi-source terrain data, including UAV lidar scanning data, satellite remote sensing image data, and ground mapping data; establishing a digital terrain model of the wind farm area based on the multi-source terrain data, dividing the terrain into terrain grid units using the digital terrain model, and constructing a multi-layered collaborative feasible domain with a ground access layer and an air flight layer based on the operating characteristics of the inspection equipment; establishing a hierarchical communication accessibility model based on the terrain grid units and identifying communication blind spots; integrating the terrain accessibility and hierarchical communication accessibility of the multi-layered collaborative feasible domain to construct a comprehensive cost field containing collaborative costs; performing inspections based on the multi-layered collaborative feasible domain and the comprehensive cost field using a two-layered collaborative optimization algorithm to generate the optimal inspection path; the two-layered collaborative optimization algorithm is used to generate a task allocation sequence based on task priority and time window constraints, and to perform path planning and smoothing correction under the constraints of the comprehensive cost field to form the optimal inspection path.

[0007] As a further improvement of the present invention, the terrain grid unit includes elevation, slope, curvature and surface roughness parameters, and the passability is judged based on the vehicle's maximum climbing angle and turning radius.

[0008] As a further improvement of the present invention, a hierarchical communication reachability model is established based on terrain grid units to identify communication blind spots, including: using a ray-crossing algorithm to calculate the line-of-sight conditions between the inspection equipment and the target based on terrain grid units, and determining the terrain occlusion situation; marking visible and invisible points according to the terrain occlusion situation, and forming a visible field partitioning model; modeling the wireless signal propagation characteristics according to the location and transmission power distribution of the communication base station, dividing local communication units in combination with the dynamic visible field partitioning model, and calculating the signal reachability of each local communication unit in real time; when the signal reachability is lower than a set threshold, the communication area is determined to be a communication blind spot, and relay node candidate points are generated at the boundary of the blind spot.

[0009] As a further improvement of the present invention, the terrain accessibility and hierarchical communication accessibility of the multi-layered collaborative feasible domain are integrated to construct a comprehensive cost field containing collaborative costs, including: acquiring multi-dimensional meteorological data, combining the slope gradient and ridge distance of the terrain grid units to construct a local wind speed distribution model; based on the local wind speed distribution model, identifying high wind speed areas, wake interference areas and turbulence-prone areas, generating a wind corridor risk distribution map and assigning corresponding risk weights; and integrating the wind corridor risk distribution, terrain accessibility and hierarchical communication accessibility data, superimposing the restricted area penalty factor to construct a comprehensive cost field.

[0010] As a further improvement of the present invention, the inspection is carried out through a two-layer collaborative optimization algorithm, including: in the two-layer collaborative optimization solution step, the first layer optimization adopts a scheduling algorithm based on task priority, the second layer optimization adopts a path planning algorithm based on constraint heuristic search, and the two-layer interactive feedback is realized through rolling iteration.

[0011] As a further improvement of the present invention, the first layer of optimization adopts a scheduling algorithm based on task priority, and the second layer of optimization adopts a path planning algorithm based on constraint heuristic search. The first layer of optimization aims at the coverage and execution timeliness of inspection tasks, generating several task allocation schemes and task sequences based on task priority and time window constraints. The second layer of optimization, under the task sequence constraints of the first layer of optimization, uses a comprehensive cost field to solve for paths. The path planning of the UAV in the inspection equipment generates an initial trajectory based on a fast travel algorithm with task sequence constraints. The path planning of the ground vehicle in the inspection equipment uses a heuristic search algorithm to generate an initial trajectory and combines vehicle dynamics constraints for curvature correction.

[0012] As a further improvement of the present invention, after obtaining the optimal inspection path, the method also includes real-time monitoring of wind speed, communication quality and energy consumption during the inspection process. When any parameter exceeds the safety threshold, an incremental path replanning algorithm is triggered to locally update the disturbed path segment in order to maintain the continuity of the task and the dynamic feasibility of the path.

[0013] Secondly, this invention provides an intelligent wind farm inspection path generation system based on a digital terrain model, comprising: a data acquisition module for acquiring multi-source terrain data, including UAV lidar scanning data, satellite remote sensing image data, and ground mapping data; a terrain modeling module for establishing a digital terrain model of the wind farm area based on the multi-source terrain data, dividing the terrain into terrain grid units using the digital terrain model, and constructing a multi-layer collaborative feasible domain with a ground access layer and an air flight layer based on the operating characteristics of the inspection equipment; a communication modeling module for establishing a hierarchical communication accessibility model based on the terrain grid units and identifying communication blind spots; and constructing a comprehensive cost field containing collaborative costs by integrating the terrain accessibility and hierarchical communication accessibility of the multi-layer collaborative feasible domain; and a path optimization module for performing inspections and generating optimal inspection paths based on the multi-layer collaborative feasible domain and the comprehensive cost field using a two-layer collaborative optimization algorithm; the two-layer collaborative optimization algorithm is used to generate a task allocation sequence based on task priority and time window constraints, and perform path planning and smoothing correction under the constraints of the comprehensive cost field to form the optimal inspection path.

[0014] Thirdly, the present invention provides a computer-readable storage medium for storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform the above-described intelligent generation method for wind farm inspection paths based on digital terrain models.

[0015] Fourthly, the present invention provides a computing device, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for performing the above-described intelligent generation method for wind farm inspection paths based on digital terrain models.

[0016] The beneficial effects of this invention are as follows: This invention provides an intelligent generation method for wind farm inspection paths based on a digital terrain model. It acquires multi-source terrain data, including UAV lidar scanning data, satellite remote sensing image data, and ground mapping data. Based on this multi-source terrain data, a digital terrain model of the wind farm area is established, and the terrain is divided into terrain grid units. According to the operating characteristics of the inspection equipment, a multi-layered collaborative feasible domain of ground accessibility and airborne flight capability is constructed, achieving a three-dimensional collaborative representation of ground and airborne accessibility in complex terrain environments. Compared with single-dimensional terrain analysis, this method has higher accessibility assessment accuracy. Furthermore, by establishing a hierarchical communication accessibility model and identifying communication blind spots, it integrates the terrain accessibility of the multi-layered collaborative feasible domain with hierarchical communication accessibility to construct a path with collaborative costs. The integrated cost field integrates the dual-dimensional constraints of terrain accessibility and communication coverage, effectively avoiding the inspection interruption problem caused by communication blind spots in traditional inspection path planning. Based on the multi-layer collaborative feasible domain and the integrated cost field, when conducting inspections through a two-layer collaborative optimization algorithm, a task allocation sequence is generated according to task priority and time window constraints. Path planning and smoothing correction are performed under the constraints of the integrated cost field. The resulting optimal inspection path satisfies the spatiotemporal constraints of the inspection task while achieving a dual improvement in inspection efficiency and safety through the collaborative optimization of the ground access layer and the air flight layer. The overall technical solution achieves the optimization and robustness enhancement of the inspection path under complex terrain and communication constraints through the synergistic effect of the multi-layer collaborative feasible domain and the integrated cost field. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 is a flowchart illustrating a method for intelligently generating wind farm inspection paths using a digital terrain model, according to an embodiment of the present invention; Figure 2 is a module diagram illustrating a system for intelligently generating wind farm inspection paths using a digital terrain model, according to an embodiment of the present invention; Figure 3 is an internal structure diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives and technical solutions of this invention clearer and easier to understand, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0020] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. The described embodiments are only some embodiments of the present invention, and not all embodiments.

[0021] Example 1 This example provides an intelligent generation method for wind farm inspection paths based on a digital terrain model. The method specifically includes: acquiring multi-source terrain data, including UAV lidar scanning data, satellite remote sensing image data, and ground mapping data; establishing a digital terrain model of the wind farm area based on the multi-source terrain data; dividing the terrain into terrain grid units using the digital terrain model; and constructing a multi-layered collaborative feasible domain with a ground access layer and an air flight layer based on the operating characteristics of the inspection equipment; establishing a hierarchical communication accessibility model based on the terrain grid units and identifying communication blind spots; integrating the terrain accessibility and hierarchical communication accessibility of the multi-layered collaborative feasible domain to construct a comprehensive cost field containing collaborative costs; and performing inspections using a two-layered collaborative optimization algorithm based on the multi-layered collaborative feasible domain and the comprehensive cost field to generate the optimal inspection path; the two-layered collaborative optimization algorithm is used to generate a task allocation sequence based on task priority and time window constraints, and to perform path planning and smoothing correction under the constraints of the comprehensive cost field to form the optimal inspection path.

[0022] The working principle of this embodiment is as follows: by acquiring multi-source terrain data, including UAV lidar scanning data, satellite remote sensing image data, and ground mapping data, more comprehensive and high-precision terrain information is provided, overcoming the incompleteness of a single data source; a digital terrain model of the wind farm area is established based on this multi-source terrain data, realizing high-resolution digital reconstruction of the terrain and laying the foundation for refined analysis; the terrain is divided into terrain grid units using the digital terrain model, providing a structured terrain representation method that facilitates quantitative processing; a multi-layered collaborative feasible domain is constructed based on the operating characteristics of the inspection equipment, including a ground access layer and an air flight layer, making path planning more aligned with the actual operating limitations of different equipment and improving practicality; a hierarchical communication reachability model is established based on the terrain grid units, and communication blind spots are identified, effectively ensuring communication continuity during the inspection process. This avoids the risk of signal interruption; by integrating the terrain accessibility and hierarchical communication accessibility of the multi-layered collaborative feasible domain, a comprehensive cost field with collaborative costs is constructed, forming a comprehensive evaluation index that simultaneously considers traffic efficiency and communication quality, thus optimizing the path evaluation criteria; based on the multi-layered collaborative feasible domain and the comprehensive cost field, a two-layered collaborative optimization algorithm is used for inspection. This algorithm generates a task allocation sequence based on task priority and time window constraints, and performs path planning and smoothing correction based on the comprehensive cost field constraints, thereby generating the optimal inspection path. This achieves the synergy of task scheduling and path optimization, significantly improving the overall efficiency and robustness of the inspection; finally, the synergistic effect of the above technical features enables ground and aerial inspection equipment to achieve multi-layered collaborative operation in complex terrain environments, achieving globally optimal inspection path planning and enhancing the intelligent level of wind farm operation and maintenance.

[0023] Example 2, as shown in Figure 1, provides an intelligent method for generating wind farm inspection paths using a digital terrain model. This method is applied to unmanned wind farm inspection scenarios under complex terrain conditions. It quantifies and models terrain features using a digital terrain model (DTM), integrating terrain constraints, communication accessibility constraints, and wind corridor risk factors to form a comprehensive cost field model. A two-layer collaborative optimization algorithm is then used to generate the optimal inspection path that conforms to terrain and environmental constraints, enabling safe, efficient, and continuous collaborative inspection between UAVs and ground inspection vehicles.

[0024] The overall process of this method includes the following five stages: 1. Terrain modeling and feasible domain construction stage: By gridding the digital terrain model of the wind farm area, the slope, roughness, curvature and elevation gradient of each grid node are calculated to construct the feasible domain of ground inspection vehicles and drones; for areas with severe terrain undulations, a slope change rate safety buffer zone is set to prevent ground vehicles from becoming unstable or skidding.

[0025] 2. Line-of-sight and communication accessibility modeling stage: Based on terrain data, establish a line-of-sight propagation model and communication accessibility distribution, identify communication blind spots, and generate a relay node set according to the communication signal attenuation law to provide link continuity constraints for path planning.

[0026] 3. Wind corridor risk modeling and multidimensional cost field construction stage: Combining wind speed data from the wind tower and nacelle, the gust acceleration factor and turbulence intensity are calculated, the risk density distribution of the wind corridor is established, and it is weighted and integrated with terrain cost, communication cost, and restricted area penalty to form a unified multidimensional comprehensive cost field model.

[0027] 4. Two-layer collaborative optimization solution stage: The upper layer uses a multi-objective optimization algorithm for task allocation and time window sorting, while the lower layer generates the optimal path based on the cost field under the dynamic constraints of UAV and ground vehicle; the joint optimization of maximizing task gain and minimizing path cost is achieved through a multi-objective genetic algorithm.

[0028] 5. Real-time execution and rolling replanning phase: During the inspection process, wind speed, communication quality and energy consumption status are monitored in real time. When any parameter exceeds the safety threshold, the incremental path replanning algorithm is triggered to locally update the disturbed path segment in order to maintain the continuity of the task and the dynamic feasibility of the path.

[0029] In summary, the method described in this embodiment achieves three-dimensional dynamic planning of wind farm inspection paths through feasible domain construction driven by digital terrain models and comprehensive cost field optimization. This method can ensure the safety of task execution, the reliability of communication links, and the optimization of energy consumption in a multi-vehicle collaborative environment, significantly improving the operational efficiency and environmental adaptability of unmanned wind farm inspections; specifically as follows: S1, Terrain Modeling and Feasible Domain Construction: In this stage, the wind farm area is spatially modeled using a digital terrain model to form the basic terrain data for the inspection task.

[0030] The system first uses UAV lidar scanning, satellite remote sensing imagery and ground mapping data to perform multi-source fusion of the wind farm's terrain elevation information, generating a digital terrain model that includes elevation, slope, curvature and surface roughness.

[0031] The model is divided into terrain grid cells, each containing location coordinates and terrain feature parameters.

[0032] Based on this, the system establishes multiple feasible domains according to the operational characteristics of inspection vehicles and drones.

[0033] The determination of the passable ground area takes into account the vehicle's maximum climbing angle, turning radius, ground adhesion coefficient, and road width restrictions; the passable air area is limited by the minimum safe height, obstacle buffer distance, and maximum operating height.

[0034] Meanwhile, to prevent discontinuities in the path at abrupt terrain changes, the system introduces slope variation constraints during the modeling process, setting risk weights for terrain units whose slope variation rate exceeds a threshold.

[0035] In this way, the feasible domain is precisely divided into a ground access layer and an air flight layer, providing a unified terrain constraint framework for subsequent path planning.

[0036] S2. Line of sight and communication accessibility modeling: Based on the terrain grid, the system establishes a line of sight and signal propagation model for the communication environment within the inspection area.

[0037] First, the ray-crossing algorithm is used to calculate the line-of-sight conditions between the drone or ground vehicle and the inspection target to determine whether there is terrain obstruction.

[0038] If no elevation in the straight-line propagation path between two points exceeds the height of the straight line, then the path point is considered visible; otherwise, it is marked as invisible.

[0039] Subsequently, the system models the propagation characteristics of wireless signals based on the deployment location and transmission power distribution of communication base stations, forming a communication reachability distribution map.

[0040] To address signal obstruction caused by the complex terrain of wind farms, this invention proposes a dynamic visible field partitioning model, which divides the wind farm into several local communication units and calculates the signal reachability of each unit in real time.

[0041] When the communication reachability is lower than the set threshold, the system automatically identifies the area as a communication blind spot and generates relay node candidate points at the boundary of the blind spot.

[0042] During the path planning phase, relay nodes can serve as auxiliary waypoints on the inspection path to ensure the continuity of communication links and the stability of data backhaul for inspection tasks.

[0043] S3. Wind corridor risk modeling and integrated cost field construction: In response to the complex wind corridor effect in the wind farm area, this stage establishes a wind corridor risk model based on the terrain model.

[0044] The system collects multi-dimensional meteorological data such as wind speed, wind direction, and turbulence intensity through wind measurement towers, cabin anemometers, and edge meteorological sensor nodes, and combines the slope gradient and ridge distance of terrain grid units to construct a local wind speed distribution model.

[0045] The model can identify areas of concentrated wind speed, areas of wake interference, and areas with high turbulence.

[0046] Based on the above results, the system generates a wind corridor risk distribution map and assigns high risk weights to high wind speed and high turbulence areas.

[0047] To achieve multi-factor coupled analysis, this invention establishes a comprehensive cost field model that integrates terrain accessibility, communication accessibility, wind corridor risk, and restricted area penalty factors.

[0048] Each factor is dynamically weighted according to the task type, enabling adaptive adjustment of costs.

[0049] For example, increase the risk weight of wind corridors in complex terrain or windy environments, and increase the weight of communication stability in long-distance inspection missions.

[0050] This comprehensive cost field is used to uniformly describe the feasibility and risk distribution of a path, providing a quantitative basis for subsequent path optimization.

[0051] S4. Two-layer collaborative optimization solution. Based on the comprehensive cost field, this invention adopts a two-layer collaborative optimization mechanism to achieve the global optimal planning of the inspection path.

[0052] The upper-level optimization module aims to improve the coverage and execution time of inspection tasks. Based on task priority and time window constraints, it generates multi-task allocation schemes and task sequences.

[0053] The lower-level optimization module uses a comprehensive cost field to solve the path under the constraints of the upper-level task sequence.

[0054] The UAV path planning uses a constrained fast travel algorithm to generate an initial trajectory and spline smoothing to ensure flight safety; the ground vehicle path planning uses a heuristic search algorithm to generate an initial trajectory and combines vehicle dynamics constraints for curvature correction.

[0055] During the path calculation process, the system monitors the strength of communication signals and wind speed fluctuations in real time. If an increase in local risk or a communication interruption trend is detected, a local path adjustment mechanism is triggered.

[0056] The optimization process achieves interactive feedback between the upper and lower layers through rolling iterations, enabling task allocation and path solving to converge in tandem, thus ensuring the overall efficiency and safety of the inspection task.

[0057] S5. Real-time execution and rolling replanning. During the inspection mission execution phase, the system continuously monitors the operational status of the drone and ground vehicles through edge computing nodes, including location, battery level, wind speed, and communication link quality.

[0058] When an environmental disturbance event is detected (such as a sudden change in wind speed, insufficient power, or communication interruption), the system triggers the rolling replanning module to perform local updates on the affected path segments without changing the overall task sequence.

[0059] The replanning module uses an incremental calculation method, adjusting only the paths in local areas to reduce computational load and improve response speed.

[0060] In communication blind spots, drones can automatically enter relay mode and maintain the communication link by increasing flight altitude or hovering to forward data; ground vehicles can then relay data with drones through near-field communication in the edge of the blind spot.

[0061] At the same time, the system will dynamically update the comprehensive cost field based on historical execution data, gradually forming a feedback self-learning mechanism.

[0062] Through multiple iterations of training, the path generation model can automatically correct the terrain feasible region, communication attenuation model, and wind farm risk weights, making the path planning results more consistent with the actual operating environment of the wind farm, thus forming an intelligent inspection system with adaptive and self-optimizing capabilities.

[0063] In this embodiment, the "system" refers to the intelligent generation system for wind farm inspection paths that combines digital terrain models to execute the method of the present invention. The system consists of a data acquisition module, a terrain modeling module, a communication modeling module, a risk assessment module, a path optimization module, an execution and feedback module, and a central control module. Its structure and connection relationship are described in detail in Embodiment 3.

[0064] In this embodiment, the system serves as the carrier for method execution, enabling computational and control functions such as terrain modeling, risk modeling, path optimization, and rolling replanning.

[0065] In summary, the intelligent generation method for wind farm inspection paths proposed in this embodiment, which combines digital terrain models, communication accessibility modeling, wind corridor risk analysis, and a two-layer collaborative optimization mechanism, achieves adaptive generation and dynamic optimization of inspection paths in complex wind farm environments.

[0066] This method first achieves high-precision modeling of the inspection area through a digital terrain model and constructs a multi-layered feasible domain based on terrain features, providing a spatial constraint basis for path generation. Then, it introduces line-of-sight and communication accessibility analysis to solve the problems of communication link breakage and unstable data backhaul in traditional path planning. On this basis, through wind corridor risk modeling and comprehensive cost field construction, it unifies and quantifies multiple factors such as terrain accessibility, meteorological disturbances, and communication stability, enabling path planning to balance safety and economy. Furthermore, it utilizes a two-layer collaborative optimization mechanism combining upper-level task allocation and lower-level path solving to form a collaborative scheduling system at the task and path levels, achieving global optimization of overall efficiency. Finally, through a rolling replanning module and an online self-learning mechanism, the system can adaptively adjust according to environmental disturbances during actual inspections, thereby ensuring the continuous feasibility and dynamic optimization of the inspection path.

[0067] The method of this invention has the following significant advantages: 1. Strong environmental adaptability: By integrating digital terrain and wind corridor models, it adapts to complex terrain and windy climate conditions; 2. High communication stability: Dynamic visual field partitioning and relay node planning effectively ensure data link continuity; 3. Intelligent path planning: A two-layer optimization and self-learning mechanism enables path self-correction and continuous optimization; 4. Low implementation cost: The method can be deployed on existing UAVs and ground inspection systems without additional hardware modifications; 5. Balancing safety and efficiency: Through multi-dimensional constraints of the comprehensive cost field, the path avoids high-risk areas while maintaining the shortest flight distance and optimal energy consumption.

[0068] Therefore, the technical solution of this embodiment realizes integrated intelligent path generation and optimization control of "terrain-communication-meteorology-task" in the complex environment of wind farms. Compared with the existing technology, it significantly improves the safety, continuity and adaptability of inspection operations, and provides basic support for building a digital and intelligent wind farm operation and maintenance system.

[0069] Example 3, as shown in Figure 2, provides an intelligent wind farm inspection path generation system that integrates a digital terrain model. This system enables intelligent planning and dynamic optimization of inspection paths under complex terrain and weather conditions. The system employs a modular architecture, including a data acquisition module, a terrain modeling module, a communication modeling module, a risk assessment module, a path optimization module, an execution and feedback module, and a central control module. These modules interact and collaborate via a system bus, forming a closed-loop system for multi-source information fusion and intelligent path generation tailored to the wind farm environment.

[0070] The system is deployed in a cloud-edge-device collaborative architecture, with the cloud responsible for task scheduling and global optimization, edge nodes undertaking real-time data calculation and path replanning, and drones and ground inspection vehicles serving as execution terminals. During the overall system operation, data acquisition, analysis and modeling, optimization and solution, and feedback control form an iterative cycle to maintain the inspection path in an optimal state in a dynamic environment. Specifically: a data acquisition module acquires multi-source terrain data, including UAV lidar scanning data, satellite remote sensing imagery, and ground mapping data; a terrain modeling module establishes a digital terrain model of the wind farm area based on the multi-source terrain data, divides the terrain into terrain grid units using the digital terrain model, and constructs a multi-layered collaborative feasible domain (DNF) with ground accessibility and airborne flight capability based on the operating characteristics of the inspection equipment; a communication modeling module establishes a hierarchical communication accessibility model based on the terrain grid units and identifies communication blind spots; it integrates the terrain accessibility and hierarchical communication accessibility of the multi-layered collaborative feasible domain to construct a comprehensive cost field containing collaborative costs; and a path optimization module performs inspections based on the multi-layered collaborative feasible domain and the comprehensive cost field using a two-layered collaborative optimization algorithm to generate the optimal inspection path. The two-layered collaborative optimization algorithm generates a task allocation sequence based on task priority and time window constraints, performs path planning and smoothing corrections under the constraints of the comprehensive cost field, and forms the optimal inspection path.

[0071] Specifically, the data acquisition module is used to acquire multi-source data within the inspection area, including terrain elevation data, meteorological data, equipment operating status data, and communication signal data.

[0072] Topographic elevation data is obtained by fusing UAV lidar with satellite remote sensing imagery, while meteorological data is collected in real time by cabin anemometers, wind towers, and ground meteorological sensor nodes.

[0073] Communication signal data is accessed by the edge gateway to check the base station signal strength and backhaul link quality.

[0074] The data acquisition module has a data synchronization and preprocessing unit, which is responsible for timestamp unification, data interpolation and noise filtering to ensure the spatiotemporal consistency of data input.

[0075] The terrain modeling module is used to build a digital terrain model (DTM) based on the collected multi-source terrain data and generate terrain grid cells.

[0076] The module includes a terrain data fusion unit and a feasible region determination unit.

[0077] The terrain data fusion unit performs spatial alignment and weighted reconstruction of data from different sources to generate a unified elevation matrix; the feasible domain determination unit performs accessibility analysis on terrain areas based on UAV flight parameters and vehicle dynamics constraints to form a multi-layered feasible domain map with ground access layer and air flight layer.

[0078] The terrain grid and feasible region information output by this module will be called by downstream modules for path planning constraints.

[0079] The communication modeling module is used to build a line-of-sight analysis and communication accessibility model for the inspection area.

[0080] This module includes a line-of-sight determination unit and a signal propagation analysis unit.

[0081] The line-of-sight determination unit calculates the visibility conditions between any two points based on the terrain elevation matrix and generates a line-of-sight matrix; the signal propagation analysis unit calculates the communication signal strength distribution map based on the base station location, transmission power, attenuation model and obstacle information.

[0082] The module generates a communication reachability grid by integrating the visual matrix and signal distribution information, and identifies communication blind spots.

[0083] At the edge of communication blind spots, the communication modeling module automatically generates candidate relay node data for use by the path optimization module.

[0084] The risk assessment module is used to identify high-risk areas within a wind farm caused by the interaction between topography and wind flow.

[0085] The module includes a wind field analysis unit and a risk quantification unit.

[0086] The wind field analysis unit establishes a wind speed and turbulence intensity distribution model by interpolating and meshing data from the wind tower and the nacelle anemometer; the risk quantification unit calculates the wind corridor risk index based on wind speed amplitude, turbulence energy and terrain slope and outputs it to the comprehensive cost field matrix.

[0087] Meanwhile, this module integrates wind field risk data with terrain feasible domain and communication accessibility grid to achieve unified measurement of multi-source risks.

[0088] The path optimization module is the core computing unit of the system, responsible for task allocation and path generation for inspection tasks.

[0089] This module includes an upper-level task scheduling unit and a lower-level path solving unit.

[0090] The upper-level task scheduling unit generates a task allocation scheme and an initial task sequence based on task priority, time window, and equipment status; the lower-level path solving unit uses the comprehensive cost field as input and employs a two-layer collaborative optimization algorithm to calculate the optimal path.

[0091] During the path optimization process, the module dynamically adjusts the communication weight and the wind corridor risk weight to achieve a multi-objective balance in path planning.

[0092] The optimization results are output to the execution module after path smoothing and curvature correction.

[0093] The execution and feedback module is used to execute the generated inspection path and realize real-time monitoring and rolling replanning.

[0094] The module includes a path execution unit, a status monitoring unit, and a rolling replanning unit.

[0095] The path execution unit controls the trajectory and speed of the UAV and ground vehicles based on the path point sequence; the status monitoring unit collects information on location, energy consumption, communication quality and wind speed changes in real time; when a risk exceeds the limit or communication is interrupted, the rolling replanning unit triggers local path adjustment to maintain mission continuity.

[0096] The execution results and environmental data are simultaneously fed back to the central control module to update the comprehensive cost field model, enabling the system to learn and continuously optimize itself.

[0097] The central control module is the core coordination unit of the system, used to manage task allocation, data flow, and model updates.

[0098] This module communicates with other modules via the system bus to coordinate the temporal relationship between terrain modeling, communication modeling, and path optimization processes.

[0099] Meanwhile, the central control module stores the feedback data into the task database and adjusts the risk weight of the wind corridor, the communication attenuation coefficient, and the terrain access threshold through the model self-learning mechanism to achieve long-term adaptive evolution of the system.

[0100] In summary, the system architecture of this embodiment achieves full-process automation from data acquisition, model building, risk analysis to path optimization and execution feedback.

[0101] Through data interaction and functional collaboration between modules, the system can generate inspection paths with high reliability and adaptability in complex terrain environments.

[0102] Compared with traditional single-layer path planning systems, this system has the following advantages: 1. Modular design, facilitating rapid deployment and expansion in different wind farms; 2. Cloud-edge collaborative computing architecture, enabling parallel optimization at the task and path levels; 3. Multi-source fusion analysis mechanism, simultaneously considering three-dimensional factors such as terrain, wind corridor, and communication; 4. Closed-loop feedback learning mechanism, achieving continuous optimization through rolling replanning and data self-learning; 5. High practicality, allowing direct integration with existing UAV swarm control systems and ground vehicle terminals.

[0103] This system, together with the method in Example 2, forms a complete closed loop of "terrain modeling - path planning - execution feedback", realizing efficient, safe and intelligent management of wind farm inspection tasks in complex environments.

[0104] In another embodiment of the present invention, a computer-readable storage medium is provided as a storage component within a terminal device, the function of which is to store programs and data. It should be noted that the computer-readable storage medium here includes not only the built-in storage component of the terminal device but also extended storage components supported by the device. Essentially, it is a tangible medium that can contain or store programs that can be invoked by or cooperate with an instruction execution system, device, or apparatus. This storage medium provides a storage area for the terminal's operating system and stores one or more instructions suitable for processor loading and execution, which can constitute one or more computer programs containing program code.

[0105] Specifically, examples of computer-readable storage media (a non-exclusive list) include: electrical connections with one or more wires, portable disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable optical disc read-only memory, optical storage devices, magnetic storage devices, or any reasonable combination of the above types.

[0106] The storage medium may also include data signals propagated as part of a baseband portion or a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any reasonable combination of both. Furthermore, computer-readable storage medium may also refer to other readable media besides conventional readable storage media, capable of sending, propagating, or transmitting programs for use or operation by an instruction execution system, apparatus, or device. Program code on the storage medium can be transmitted via any suitable medium, including but not limited to wireless, wired, optical fiber, or any reasonable combination thereof.

[0107] The program code used to implement the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C. The execution modes of the program code include: running entirely on the user's computing device, running partially on the user's device as a standalone software package, running partially in a distributed manner on both the user's device and a remote computing device, or running entirely on a remote computing device or server. When a remote computing device is involved, the device can be connected to the user's computing device via any type of network such as a local area network (LAN) or a wide area network (WAN), or connected to an external computing device via the Internet through an Internet service provider.

[0108] The processor is capable of loading and executing one or more instructions stored in a computer-readable storage medium to implement the corresponding steps of the intelligent generation method for wind farm inspection paths based on digital terrain models described in Example 1.

[0109] Figure 3 is a schematic diagram of a computer device provided in an embodiment of the present invention.

[0110] Please refer to Figure 3. The terminal device is a computer device. The computer device 60 in this embodiment includes a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the processor 61 executes the computer program 63, it implements the intelligent generation method for wind farm inspection paths based on digital terrain models in this embodiment. To avoid repetition, it will not be described in detail here. Alternatively, when the processor 61 executes the computer program 63, it implements the functions of each model / unit in the wind farm inspection path composition computing system based on digital terrain models in this embodiment. To avoid repetition, it will not be described in detail here.

[0111] Computer device 60 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. Computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will understand that Figure 3 is merely an example of computer device 60 and does not constitute a limitation on computer device 60. It may include more or fewer components than illustrated, or combine certain components, or use different components. For example, computer device may also include input / output devices, network access devices, buses, etc.

[0112] The processor 61 may be a central processing unit (CPU), or other general-purpose processors, CPUs, graphics processing units (GPUs), 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, quantum computing-based data processing logic units, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0113] The memory 62 can be an internal storage unit of the computer device 60, such as a hard disk or RAM of the computer device 60. The memory 62 can also be an external storage device of the computer device 60, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the computer device 60.

[0114] Furthermore, the memory 62 may include both internal storage units of the computer device 60 and external storage devices. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 can also be used to temporarily store data that has been output or will be output.

[0115] Any references to memory, databases, or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (Read-Only Memory). Memory includes ROM, magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0116] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

Claims

1. A method for intelligent generation of wind farm inspection paths based on digital terrain models, characterized in that, include: Acquire multi-source terrain data, including UAV lidar scanning data, satellite remote sensing image data, and ground mapping data; A digital terrain model of the wind farm area is established based on multi-source terrain data. The terrain is divided into terrain grid units using the digital terrain model. A multi-layer collaborative feasible domain of ground access layer and air flight layer is constructed based on the operating characteristics of the inspection equipment. A hierarchical communication reachability model is established based on terrain grid cells, and communication blind spots are identified. By integrating the terrain accessibility and hierarchical communication reachability of the multi-layered collaborative feasible domain, a comprehensive cost field with collaborative costs is constructed. Based on the multi-layered collaborative feasible domain and the comprehensive cost field, a two-layered collaborative optimization algorithm is used to perform inspection and generate the optimal inspection path. The two-layered collaborative optimization algorithm is used to generate a task allocation sequence based on task priority and time window constraints, and to perform path planning and smoothing correction under the constraints of the comprehensive cost field to form the optimal inspection path.

2. The intelligent generation method for wind farm inspection paths based on digital terrain models according to claim 1, characterized in that, The terrain grid unit includes parameters such as elevation, slope, curvature, and surface roughness, and its passability is determined based on the vehicle's maximum climbing angle and turning radius.

3. The intelligent generation method for wind farm inspection paths based on digital terrain models according to claim 1, characterized in that, Based on terrain grid units, a hierarchical communication reachability model is established and communication blind spots are identified. This includes: using a ray-crossing algorithm based on terrain grid units to calculate the line-of-sight conditions between the inspection equipment and the target, and determining the terrain occlusion situation; marking visible and invisible points according to the terrain occlusion situation, and forming a visible field partitioning model; modeling the wireless signal propagation characteristics according to the location and transmission power distribution of communication base stations, dividing local communication units in combination with the dynamic visible field partitioning model, and calculating the signal reachability of each local communication unit in real time; when the signal reachability is lower than a set threshold, the communication area is determined to be a communication blind spot, and relay node candidate points are generated at the boundary of the blind spot.

4. The intelligent generation method for wind farm inspection paths based on digital terrain models according to claim 3, characterized in that, By integrating the terrain accessibility and hierarchical communication accessibility of the multi-layered collaborative feasible domain, a comprehensive cost field with collaborative costs is constructed, including: acquiring multi-dimensional meteorological data, combining the slope gradient and ridge distance of the terrain grid units to construct a local wind speed distribution model; based on the local wind speed distribution model, identifying high wind speed areas, wake interference areas, and turbulence-prone areas, generating a wind corridor risk distribution map and assigning corresponding risk weights; and integrating the wind corridor risk distribution, terrain accessibility, and hierarchical communication accessibility data, superimposing the restricted area penalty factor, and constructing a comprehensive cost field.

5. The intelligent generation method for wind farm inspection paths based on digital terrain models according to claim 1, characterized in that, Inspection is performed using a two-layer collaborative optimization algorithm, including: in the two-layer collaborative optimization solution steps, the first layer of optimization adopts a scheduling algorithm based on task priority, the second layer of optimization adopts a path planning algorithm based on constraint heuristic search, and the two-layer interactive feedback is realized through rolling iteration.

6. The intelligent generation method for wind farm inspection paths based on digital terrain models according to claim 5, characterized in that, The first-level optimization employs a task priority-based scheduling algorithm, while the second-level optimization employs a constraint-based heuristic search-based path planning algorithm. This includes: the first-level optimization aims to maximize the coverage and execution time of inspection tasks, generating several task allocation schemes and task sequences based on task priority and time window constraints; the second-level optimization, under the task sequence constraints of the first-level optimization, utilizes a comprehensive cost field for path solving; the path planning for UAVs in the inspection equipment generates initial tracks based on a task sequence constraint-based fast travel algorithm; and the path planning for ground vehicles in the inspection equipment uses a heuristic search algorithm to generate initial tracks, combined with vehicle dynamics constraints for curvature correction.

7. The intelligent generation method for wind farm inspection paths based on digital terrain models according to claim 1, characterized in that, After obtaining the optimal inspection path, the process also includes real-time monitoring of wind speed, communication quality, and energy consumption during the inspection. When any parameter exceeds the safety threshold, an incremental path replanning algorithm is triggered to locally update the disturbed path segment in order to maintain the continuity of the task and the dynamic feasibility of the path.

8. A wind farm inspection path intelligent generation system based on digital terrain model, characterized in that, include: The data acquisition module is used to acquire multi-source terrain data, including UAV lidar scanning data, satellite remote sensing image data, and ground mapping data. The terrain modeling module establishes a digital terrain model of the wind farm area based on multi-source terrain data. It uses the digital terrain model to divide the terrain into terrain grid units and constructs a multi-layer collaborative feasible domain of ground access layer and air flight layer based on the operating characteristics of inspection equipment. The communication modeling module is used to build a hierarchical communication accessibility model and identify communication blind spots based on terrain grid cells; By integrating the terrain accessibility and hierarchical communication reachability of the multi-layered collaborative feasible domain, a comprehensive cost field containing collaborative costs is constructed; the path optimization module is used to perform inspections based on the multi-layered collaborative feasible domain and the comprehensive cost field, and generate the optimal inspection path through a two-layer collaborative optimization algorithm. The two-layer collaborative optimization algorithm is used to generate a task allocation sequence based on task priority and time window constraints, and to perform path planning and smoothing correction under the constraint of comprehensive cost field to form the optimal inspection path.

9. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the intelligent generation method for wind farm inspection paths based on a digital terrain model as described in any one of claims 1 to 7.

10. A computing device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including steps for performing the steps of the intelligent generation method for wind farm inspection paths based on a digital terrain model according to any one of claims 1 to 7.

Images