Unmanned aerial vehicle fan inspection route autonomous planning method

Abstract

The invention discloses an unmanned aerial vehicle fan inspection route autonomous planning method, and relates to the technical field of unmanned aerial vehicle inspection, and the method comprises the following specific steps: real-time data acquisition: establishing a data sharing link with a professional marine monitoring mechanism to obtain tide data, deploying local monitoring equipment in an offshore fan group area, and establishing a data sharing link; according to the method, through fusion of dynamic tide compensation, tide prediction and real-time monitoring data can be obtained in real time, and a dynamic marine environment model is constructed, so that marine environment conditions at different tide moments are accurately reflected; therefore, the unmanned aerial vehicle can avoid a risk area where seawater rises and floaters are increased during flood tide, select a proper flight height and ensure safety, and can be close to a fan foundation for detailed inspection during ebb tide, and meanwhile, a safety buffer area is set by utilizing dangerous area information in the model, so that the inspection safety is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of drone inspection, and specifically to an autonomous planning method for the inspection route of drones for wind turbines. Background Art

[0002] As an important part of renewable energy, the offshore wind turbine group is of great significance for ensuring power supply. To ensure the stable operation and timely maintenance of offshore wind turbines, the inspection work is particularly important. As an efficient and low-cost inspection method, drone inspection has been widely used in the inspection of offshore wind turbine groups in recent years. However, the inspection work of offshore wind turbine groups faces complex and changeable working conditions, especially factors such as tidal changes and sudden wind turbine failures, which pose great challenges to drone inspection.

[0003] Traditional drone inspection route planning methods often do not fully consider the impact of tidal dynamic changes on the marine environment. The periodic ebb and flow of tides will change the marine environment. When the tide rises, the sea water rises and the floating objects increase, and drones may face safety risks during the inspection process. When the tide ebbs, new dangerous areas may appear, such as exposed reefs and shoals. In addition, when a sudden wind turbine failure occurs, the traditional inspection system cannot quickly and reasonably adjust the inspection route, and does not incorporate the tidal situation into the emergency route planning, resulting in a serious impact on the timeliness and effectiveness of fault handling. Therefore, traditional drone inspection route planning methods have obvious deficiencies in dealing with the complex working conditions of offshore wind turbine groups.

[0004] In view of the above problems, it is necessary to optimize the existing drone inspection route planning method for wind turbines. By comprehensively considering the tidal dynamic changes and sudden wind turbine failures, the intelligent planning of the drone inspection route is realized. Therefore, it is of great significance to develop an autonomous planning method for the inspection route of drones for wind turbines that can comprehensively achieve the above characteristics. Summary of the Invention

[0005] The purpose of the present invention is to make up for the deficiencies of the existing technology and provide an autonomous planning method for the inspection route of drones for wind turbines. It can construct a dynamic marine environment model by real-time obtaining tidal data and the operating status data of wind turbines, and on this basis, plan the basic inspection route. At the same time, this method also has the function of monitoring and warning sudden wind turbine failures, and can immediately start the emergency route replanning program when a failure occurs, ensuring that the drone can still efficiently and safely complete the inspection task under complex working conditions.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an autonomous planning method for the inspection route of drones for wind turbines, which includes the following specific steps:

[0007] Real-time data acquisition: We acquire tidal data by establishing data sharing links with professional marine monitoring institutions, deploying local monitoring equipment in the offshore wind turbine cluster area, and building multiple acquisition channels to acquire accurate tidal data. At the same time, we deploy vibration, temperature, speed and image sensors on wind turbines to collect real-time data on wind turbine location, structural characteristics and operating status;

[0008] Construction of dynamic offshore environment model: Use geographic information system technology to digitally model the location and structure information of wind turbines, generate a spatial layout model of offshore wind turbines, combine tidal data, use numerical simulation methods to simulate the dynamic changes of seawater with the help of hydrodynamic models, identify dangerous areas and integrate them into the model, and build a dynamic model that can reflect the overall picture of the offshore environment in real time;

[0009] Basic inspection route planning: Based on the dynamic marine environment model, the path planning algorithm is used to plan inspection routes that cover key parts of wind turbines and avoid dangerous areas in response to rising sea water and increasing floating objects during high tide. The inspection routes are planned based on the layout and orientation of wind turbines. In view of the exposed characteristics of wind turbine foundations during low tide, the flight altitude is adjusted to plan inspection routes that are close to the wind turbine foundations and keep a safe distance from dangerous areas.

[0010] Monitoring and early warning of sudden faults of wind turbines: When the wind turbine is running, vibration, temperature, speed and image data are collected in real time and transmitted to the data analysis model. The fault diagnosis model is used to deeply analyze the data. When an abnormality is identified, the wind turbine is determined to be faulty, thereby triggering an early warning, and the fault information is transmitted to the route planning system and operation and maintenance personnel;

[0011] Emergency route re-planning: When a fault warning signal is received, the current tidal data is obtained to determine the tidal stage, safe flight altitude range and affected area. Based on the status of the faulty wind turbine and surrounding wind turbines, real-time tidal data, and through optimization algorithms, the impact of the tide on the surrounding area is fully considered and the existing inspection routes are comprehensively adjusted.

[0012] Furthermore, in the step of constructing the dynamic marine environment model, in combination with tidal data, a numerical simulation method is used to simulate the dynamic changes of seawater with the help of a hydrodynamic model. Specifically, the influence of the topography and coastline shape of the sea area on the flow of seawater is considered, and the changes in the coverage of seawater during high tide and low tide are simulated. The coverage of seawater is determined through multiple iterative calculations, and the influence of ocean dynamics principles and astronomical factors on tides is comprehensively considered to simulate the tidal flow rate and direction at different locations and times. In addition, through the analysis of historical tidal data and seabed topography data, the dangerous areas of reefs and shoals exposed after low tide are determined, and this information is integrated into the dynamic model.

[0013] Furthermore, in the basic inspection route planning step, according to the dynamic marine environment model, in view of the rising sea water and increasing floating objects during high tide, the flight altitude range is determined by comprehensively considering the tide height, the distribution of floating objects and the flight performance of the UAV. The path planning algorithm is used to cover the cabin, blades and tower parts as the goal, combined with the layout and orientation of the wind turbines, and the flight distance and time cost are reduced by optimizing the path sequence and turning angle. At the same time, according to the information of the sea water and floating object distribution during high tide in the model, the dangerous area is set as an impassable obstacle to avoid, and the route planning strategy is adjusted according to the flight data feedback by the UAV in real time.

[0014] Furthermore, in the basic inspection route planning step, the path planning algorithm is used, and its algorithm formula is: Among them, C ZC is the comprehensive cost of the candidate route in the high tide stage, k is the number of key parts of the wind turbine passed by the route, and α, β, and γ are the weight coefficients of the dangerous distance, flight altitude, and coverage degree of the key parts of the wind turbine respectively, which are determined by using the analytic hierarchy process through the analysis of historical inspection data and accident data. is the distance between the route and the j-th key part of the wind turbine and the nearest dangerous area, d max is the maximum distance between the dangerous area and the key part of the wind turbine in the current high tide scenario, which is determined by the dynamic marine environment model combined with historical data and geographical information, h j is the flight altitude of the UAV reaching the j-th key part of the wind turbine, which is determined by the flight altitude range set according to the high tide situation, h AQ is the safe flight altitude, which is determined according to the tide and floating object situation, p j is the inspection coverage degree of the route for the j-th key part of the wind turbine, which is calculated through the relationship between the preset inspection points of the key parts of the wind turbine and the route.

[0015] Furthermore, in the basic inspection route planning step, in view of the characteristics of the wind turbine foundation being exposed during ebb tide, the flight altitude is adjusted. Specifically, the planning algorithm is used to plan the inspection route close to the wind turbine foundation around each inspection point of the wind turbine foundation structure, combined with its shape, size and inspection point distribution. At the same time, with the help of the information of the dangerous areas including reefs and shoals after ebb tide in the model, according to the degree of danger, the flight speed and maneuverability of the UAV, a safety buffer zone is set around the route and regarded as an impassable area to ensure safety. At the same time, the inspection route is adjusted according to the characteristics of the wind turbine foundation structure and the important inspection parts.

[0016] Furthermore, in the basic inspection route planning step, the path planning algorithm is used to plan the inspection route close to the wind turbine foundation around each inspection point of the wind turbine foundation structure, combined with its shape, size and inspection point distribution. Its algorithm formula is: Among them, C TCis the comprehensive cost of the candidate route in the low tide stage. The lower the cost, the better the route. l is the number of wind turbine basic inspection points along the route. δ and ∈ are the weight coefficients of the distance to the dangerous area and the coverage of wind turbine basic inspection, respectively. By analyzing the historical low tide inspection data and accident data, and taking each inspection reward as feedback, δ and ∈ are adjusted to make the route adapt to environmental changes. is the distance between the route and the m wind turbine foundation inspection points and the nearest dangerous area, determined by the dynamic offshore environment model, d AQJL The safe distance is determined according to the degree of danger, which is determined by the marine environment model combined with historical data. m is the inspection detail level of the route for the mth wind turbine foundation inspection point, which is calculated based on the importance of the wind turbine foundation inspection point and the proximity between the route and the inspection point.

[0017] Furthermore, in the wind turbine sudden fault monitoring and early warning step, a fault diagnosis model is used to deeply analyze the data, and the model formula is: Among them, G is the fault judgment index. When G is greater than the threshold G max When the fan fails, it is determined that a fault has occurred and an early warning is triggered. n represents the number of data types collected, and w i is the weight coefficient of the i-th data type, which is determined by analyzing historical fault data and machine learning technology. i (x i ) is the characteristic function of the i-th data type, which is used to convert the collected raw data x i It is converted into a characteristic value that can reflect the possibility of fault. b is the weight coefficient of the comprehensive characteristics of image data, which is determined according to the importance and reliability of image data in fault judgment. g(y) is the comprehensive characteristic function of image data, and y represents the collected image data.

[0018] Furthermore, in the emergency route re-planning step, the influence of tides on the surrounding areas is fully considered through optimization algorithms, and the existing inspection routes are comprehensively adjusted. Specifically, the tide height and tide rise and fall status data are retrieved in real time, and the location, type, severity of the faulty wind turbine and the real-time operating status of the surrounding wind turbines are collected. Based on these data, the fitness function is used to evaluate the candidate emergency routes, and candidate emergency routes are generated according to the faulty wind turbine, surrounding wind turbines and tidal impact areas. Genetic operations are performed on the candidate routes, and individuals are selected through the roulette selection method. After multiple generations of genetic operations, the route corresponding to the individual with the best fitness is selected as the emergency route. If the fitness value does not increase significantly for multiple consecutive generations, the genetic algorithm parameters are adjusted, and the emergency route is sent to the drone to perform the inspection task.

[0019] Furthermore, in the emergency route replanning step, a fitness function is used to evaluate the candidate emergency routes, and the function formula is: Among them, F is the fitness value of the candidate route, which is used to evaluate the quality of the candidate route. T is the time required for the UAV to reach the faulty wind turbine, which is obtained by calculating the route length and the UAV flight speed. The route length is calculated based on the coordinates of each point on the path. λ is the distance influence factor, which is adjusted according to the urgency of the fault and the importance of the surrounding wind turbines. D is the sum of the distances between the route and other wind turbines, which is obtained by calculating and accumulating the distances between the route and the positions of the surrounding wind turbines. μ is the risk influence factor, which is adjusted according to the danger level under the current tidal conditions. R is the route risk value, which is determined by comprehensively considering the dangerous area under the influence of tides and the possible dangerous range generated by the faulty wind turbine.

[0020] Furthermore, in the emergency route replanning step, individuals are selected by the roulette wheel selection method, and its formula is: Where P i represents the probability that the i-th candidate emergency route is selected for the next generation of reproduction. F i is the fitness value of the i-th candidate emergency route. N represents the population size, that is, the total number of candidate emergency routes that participate in the iterative optimization of the genetic algorithm during the emergency route replanning process. F j is the fitness value of the j-th candidate emergency route.

[0021] Compared with the prior art, the UAV wind turbine inspection route autonomous planning method has the following beneficial effects:

[0022] First, by integrating dynamic tide compensation, the present invention can obtain real-time tide prediction and real-time monitoring data, construct a dynamic offshore environment model, so as to accurately reflect the offshore environmental conditions at different tidal moments. Furthermore, the UAV can avoid the risk areas where the sea water rises and the floating objects increase during high tide, select an appropriate flight altitude to ensure safety, and can conduct a detailed inspection close to the wind turbine foundation during low tide. At the same time, a safety buffer zone is set using the dangerous area information in the model, further improving the safety of the inspection.

[0023] Second, by installing fault monitoring sensors, the present invention constructs a wind turbine operation status monitoring network, real-time monitors the operation status of the wind turbines, and immediately triggers a warning signal when an abnormality is identified. An intelligent optimization algorithm is used to adjust the existing inspection route to ensure that the UAV can quickly reach the position of the faulty wind turbine for inspection. At the same time, considering the influence of tides on the surrounding area, the inspection tasks of other UAVs are reasonably adjusted, so as to maximize the use of UAV resources while ensuring the inspection efficiency, and enhance the timeliness and effectiveness of emergency response.

[0024] Other advantages, objectives and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. Description of the Drawings

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 It is a flowchart of an autonomous planning method for the inspection route of an unmanned aerial vehicle (UAV) fan.

[0027] Figure 2 It is a flow operation diagram of an autonomous planning method for the inspection route of an unmanned aerial vehicle (UAV) fan. Detailed Embodiments

[0028] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, detail the specific embodiments, structures, features, and their effects of the present invention as follows.

[0029] Embodiment 1

[0030] By establishing a data dedicated line with a professional ocean monitoring institution, wide-area tidal data is obtained in real time. At the same time, 10 sets of local high-precision tidal monitoring stations are evenly deployed in the wind farm, and data such as tidal level, tidal current velocity, flow direction, and flood and ebb tide times are collected every 5 minutes. For example, before a typhoon hit, through data sharing and local monitoring, abnormal tidal changes were grasped in advance, providing data support for subsequent operations. Vibration, temperature, rotation speed sensors, and high-definition cameras are respectively installed at positions such as the blade root, hub, key components inside the nacelle, and tower barrel of each fan. Each type of sensor collects data in real time at a frequency of 100 Hz and quickly transmits it to the central data processing platform through the 5G network. For example, the blade vibration sensor can accurately capture the vibration amplitude and frequency of the blade under different working conditions, providing detailed data for fault diagnosis.

[0031] Using high-precision GIS technology, accurately digitize the three-dimensional spatial position, foundation structure size, tower height, etc. of the wind turbines to generate a detailed spatial layout model of the offshore wind turbine group, accurate to the centimeter level. Use advanced numerical simulation software, combined with real-time tidal data and ocean hydrodynamic models, to accurately simulate the dynamic changes such as the coverage area, flow velocity, flow direction, and wave height of seawater at different tidal stages. Specifically, consider the influence of the terrain and coastline shape factors of the sea area on the seawater flow, simulate the change in the coverage area of seawater during the flood tide and ebb tide processes, determine the coverage area of seawater through multiple iterative calculations, and comprehensively consider the principles of ocean dynamics and the influence of astronomical factors on tides to simulate the tidal current velocity and flow direction at different positions and times. In addition, through the analysis of historical tidal data and seabed terrain data, determine the dangerous areas of reefs and shoals exposed after the ebb tide, and integrate this information into the dynamic model.

[0032] For the flood tide stage, based on the dynamic offshore environment model, comprehensively consider the distribution of floating objects, seawater height, and the performance of the unmanned aerial vehicle (UAV) to determine that the safe flight height range is 60 - 90 meters. Use a path planning algorithm with the goal of covering key parts such as the wind turbine blades, nacelle, and tower, while avoiding dangerous areas. During this process, evaluate the candidate routes through the formula where C ZC is the comprehensive cost of the candidate route during the flood tide stage, k is the number of key parts of the wind turbine passed by the route, and α, β, and γ are the weight coefficients of the dangerous distance, flight height, and coverage degree of the key parts of the wind turbine respectively. Initially, they are determined by using the analytic hierarchy process through the analysis of historical inspection data, accident data, and expert experience, and α + β + γ = 1. For example, at the initial stage of the project, based on past experience and data analysis, initially set α = 0.4, β = 0.3, γ = 0.3. As the inspection data accumulates continuously, if during a certain inspection, the key parts of the wind turbine blades cannot be clearly observed due to the flight height problem, missing potential faults, then appropriately increase β to 0.4, reduce γ to 0.2, while keeping α = 0.4. If during subsequent inspections, approaching a dangerous area triggers a safety alarm, then increase α to 0.5, β = 0.3, γ = 0.2. is the distance from the route to the j-th key part of the wind turbine and the nearest dangerous area, determined by the offshore dynamic model, and d max is the maximum distance between the dangerous area and the key parts of the wind turbine in the current flood tide scenario, determined by the dynamic offshore environment model combined with historical data and geographical information, and h j is the flight height of the UAV reaching the j-th key part of the wind turbine, determined by the flight height range set according to the flood tide situation, and h AQ is the safe flight height, determined according to the tide and the situation of floating objects, and p j$P_{j}$ is the inspection coverage of the $j$-th key part of the wind turbine by the flight path, which is calculated based on the relationship between the preset inspection points of the key parts of the wind turbine and the flight path. For example, during a flood tide inspection, an efficient flight path is successfully planned. The UAV flies along the flight path, taking comprehensive and clear photos and collecting data of all key parts of the wind turbine while avoiding the floating object aggregation area. During the ebb tide stage, according to the model feedback, the flight altitude is adjusted to 25 - 40 meters, and an inspection flight path close to the wind turbine foundation is planned using an algorithm. An 8-meter-wide safety buffer zone is set around the flight path to avoid the dangerous areas exposed after the ebb tide. At this time, the formula is used to evaluate the candidate flight path, where $C$ TC is the comprehensive cost of the candidate flight path during the ebb tide stage. The lower the cost, the better the flight path. $l$ is the number of inspection points of the wind turbine foundation passed by the flight path. $\delta$ and $\epsilon$ are the weight coefficients of the distance to the dangerous area and the inspection coverage of the wind turbine foundation respectively. Initially, it is determined by analyzing historical ebb tide inspection data and accident data, and $\delta+\epsilon = 1$. For example, at the beginning of the project, $\delta = 0.6$ and $\epsilon = 0.4$ are initially determined. If a loose bolt fault of the wind turbine foundation is not found during an ebb tide inspection due to low inspection coverage, then $\epsilon$ is increased to 0.5 and $\delta$ is decreased to 0.5. If a potential risk is caused by approaching a dangerous area during subsequent inspections, $\delta$ is increased to 0.7 and $\epsilon = 0.3$. is the distance between the flight path and the $m$ inspection points of the wind turbine foundation and the nearest dangerous area, which is determined by the dynamic offshore environment model. $d$ AQJL is the safety distance determined according to the degree of danger, which is determined by the dynamic offshore environment model combined with historical data. $s$ m is the inspection detail level of the $m$-th wind turbine foundation inspection point by the flight path, which is calculated based on the importance of the wind turbine foundation inspection point and the proximity of the flight path to the inspection point. During an ebb tide inspection, the UAV conducted a detailed inspection of the wind turbine foundation along the planned flight path, including the fastening condition of the foundation bolts and the cracks on the concrete surface.

[0033] During the operation of the wind turbine, vibration, temperature, rotation speed, and image data are collected in real time and transmitted to the data analysis model. A fault diagnosis model is used to deeply analyze the data, and a fault mode recognition system is constructed through learning and training. When an abnormality is identified and a fault is determined, an early warning is immediately triggered, and the fault location, type, and severity information are transmitted to the flight path planning system and the operation and maintenance personnel. According to the fault early warning signal, the current tide data and fault-related information are obtained. For example, if the fault occurs in the late flood tide stage, the seawater flow velocity is relatively fast, and the dangerous area expands. Based on the location of the faulty wind turbine, the type of fault, the status of surrounding wind turbines, and the tide data, the fitness function Evaluate candidate emergency routes, where F is the fitness value of an individual (candidate route), and the higher the value, the better the route. T is the time required for the UAV to reach the faulty wind turbine, which is obtained by calculating the route length and flight speed. The route length is calculated by accumulating the distances between points on the path using the distance formula between two points. λ is the distance influence factor, which is initially set according to the severity of the fault and the importance of surrounding wind turbines. For example, if the fault is a blade break and the situation is urgent, λ is initially set to 0.6. If during subsequent emergency inspections, it is found that the UAV approaches a dangerous area due to excessive focus on distance and neglect of wind conditions, λ is appropriately reduced to 0.5. μ is the risk influence factor, which is initially set according to the danger level under the current tidal conditions. For example, if the dangerous area expands during the late flood tide, μ is initially set to 0.4. If during subsequent emergency inspections, the time to reach the faulty wind turbine is too long due to excessive risk consideration, μ is appropriately reduced to 0.3. D is the sum of the distances between the route and other wind turbines, which is calculated. R is the route risk value, which is determined by comprehensively considering factors such as the dangerous area affected by tides and the possible dangerous range of the faulty wind turbine. For example, if there are a large number of floating objects near the faulty wind turbine, the value of R is increased to guide the planning of a safer route, and through the roulette wheel selection method Select individuals, perform crossover and mutation operations. After multiple generations of genetic operations, determine the optimal emergency route, where P i is the probability that the i-th individual (candidate route) is selected for reproduction in the next generation, F i is the fitness value of the i-th individual, and N is the population size. In an emergency route planning for a blade fault, after 501 generations of genetic operations, a route that can quickly reach the faulty wind turbine and avoid dangerous areas is obtained. The UAV quickly reaches the fault site according to this route, providing timely and accurate fault information for maintenance personnel.

[0034] Example 2

[0035] By reaching a cooperation agreement with local marine observation institutions and leveraging data transmission lines, real-time and stable acquisition of tidal data is achieved. At the same time, fully considering the geographical characteristics of the region, 3 sets of local tidal monitoring devices are deployed in the project sea area. These devices are equipped with high-precision sensors that can accurately measure key parameters such as tide level, tidal current velocity, and flow direction. And the data collected every 10 minutes is transmitted to the data processing center through a wireless communication module. Using lightweight GIS technology, combined with the detailed geographical information of the sea area around the island and parameters such as the actual location, foundation shape, and tower height of the wind turbines, an accurate spatial layout model of the offshore wind turbine group is constructed. This model can not only clearly show the specific location of each wind turbine in the sea area but also present in detail the shape, size of the wind turbine foundation, and the height of the tower, providing an intuitive and accurate spatial reference for subsequent route planning. Combining real-time tidal data and a simplified marine hydrodynamic model specifically developed for this region, the changes in the seawater coverage area, velocity, and flow direction are accurately simulated. During the simulation process, the blocking and diversion effects of the island terrain on seawater flow, as well as the special influence of local ocean currents, are fully considered. For example, the model can simulate the complex circulation area formed by seawater after being blocked by the island and the subtle changes in velocity and flow direction during a specific tidal stage. Through this simulation, the dynamic change trend of seawater can be predicted in advance, providing strong support for route planning and helping planners avoid potential dangerous areas. At the same time, by deeply analyzing the historical tidal data and marine meteorological data of this region in the past 5 years and combining the results of multiple on-site surveys, a comprehensive understanding of the topography and geomorphology around the island is obtained. On this basis, the dangerous areas such as reefs and shoals that may appear after ebb tide are accurately determined and marked in the dynamic model.

[0036] For the flood tide stage, based on the constructed dynamic offshore environment model, comprehensively considering the rising height of seawater during flood tide, the distribution of floating objects, and the flight performance of the UAV itself, the safe flight altitude range is determined to be 40 - 70 meters. Using a path planning algorithm, combined with the actual geographical conditions and the distribution characteristics of wind turbines in this area, with the core goal of comprehensively covering key parts such as the blades, nacelles, and towers of wind turbines, while ensuring that the flight path can effectively avoid dangerous areas. For example, when planning the flight path, fully consider the possible areas where floating objects may gather during flood tide, and through algorithm optimization, make the flight path as far away from these dangerous areas as possible, while ensuring a comprehensive inspection of the key parts of the wind turbines. For the ebb tide stage, according to the information feedback from the dynamic model, considering that the wind turbine foundation is relatively exposed after ebb tide and there may be dangers such as reefs around it, the flight altitude is adjusted to 15 - 30 meters to inspect the wind turbine foundation at a closer distance. Use the algorithm to plan a flight path close to the wind turbine foundation. During the planning process, fully consider the shape, size of the wind turbine foundation and the distribution of surrounding dangerous areas, and set a 5 - meter - wide safety buffer zone for the flight path to ensure that the UAV can effectively avoid dangers when approaching the wind turbine foundation for inspection. For example, in a certain ebb tide inspection, through the flight path planned by this algorithm, the UAV can conduct a comprehensive inspection of the wind turbine foundation within a safe distance, including the stability of the foundation and the integrity of the connection components.

[0037] During the operation of the fan, vibration, temperature, rotational speed and image data are collected in real time and transmitted to the data analysis model. The fault diagnosis model is used to deeply analyze the data, and a fault mode recognition system is constructed through learning and training. When an anomaly is recognized and a fault is determined, an early warning is immediately triggered, and information about the fault location, type and severity is transmitted to the route planning system and the operation and maintenance personnel. According to the fault early warning signal, the current tidal data, including key information such as tidal level height, flood and ebb tide states, tidal current velocity and direction, is quickly retrieved from the real-time data acquisition module. At the same time, data such as the location of the faulty fan, the fault type, the fault severity, and the real-time operating status of the surrounding fans are comprehensively collected. Based on the obtained location of the faulty fan, the fault type, the status of the surrounding fans, and the tidal data, the route planning system will initiate an emergency route planning process. First, according to the urgency of the fault and the complexity of the surrounding environment, the focus and direction of the planning are determined. For example, for an emergency fault such as a blade breakage, the path to quickly reach the faulty fan is given priority. At the same time, the impact of the tide on the surrounding area is fully considered, such as dangerous areas such as reefs and shoals exposed during ebb tide, and the impact of the seawater coverage range and velocity during flood tide on the flight of the UAV. During the planning process, intelligent algorithms are used to evaluate and screen multiple candidate routes, comprehensively considering factors such as the length of the route, the distance from dangerous areas, and the impact on surrounding fans, to ensure that the planned emergency route can reach the faulty fan quickly and ensure the flight safety of the UAV. After initially planning multiple candidate emergency routes, these routes are further optimized. After multiple rounds of optimization and comparison, the optimal emergency route is determined. For example, in an emergency route planning for a blade fault, several routes that may be close to dangerous areas are excluded through simulated flights and risk assessments of multiple candidate routes. Finally, a route that can quickly reach the faulty fan, avoid dangerous areas, and have the least impact on surrounding fans is determined. The UAV quickly reaches the fault site according to this route, provides accurate fault information for the maintenance personnel, helps the maintenance personnel quickly formulate a maintenance plan, and restores the normal operation of the fan, ensuring the stability of the power supply on the island.

[0038] The above is only a preferred embodiment of the present invention and does not impose any form of limitation on the present invention. Although the present invention has been disclosed above with the preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the above-disclosed technical content to be equivalent embodiments with equivalent changes, but as long as the technical content of the present invention is not departed from, any brief modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. The autonomous planning method of the UAV wind turbine inspection route is characterized by: The system consists of the following components: Real-time data acquisition: We acquire tidal data by establishing data sharing links with professional marine monitoring institutions, deploying local monitoring equipment in the offshore wind turbine cluster area, and building multiple acquisition channels to acquire accurate tidal data. At the same time, we deploy vibration, temperature, speed and image sensors on wind turbines to collect real-time data on wind turbine location, structural characteristics and operating status; Construction of dynamic offshore environment model: Use geographic information system technology to digitally model the location and structure information of wind turbines, generate a spatial layout model of offshore wind turbines, combine tidal data, use numerical simulation methods to simulate the dynamic changes of seawater with the help of hydrodynamic models, identify dangerous areas and integrate them into the model, and build a dynamic model that can reflect the overall picture of the offshore environment in real time; Basic inspection route planning: Based on the dynamic marine environment model, the path planning algorithm is used to plan inspection routes that cover key parts of wind turbines and avoid dangerous areas in response to rising sea water and increasing floating objects during high tide. The inspection routes are planned based on the layout and orientation of wind turbines. In view of the exposed characteristics of wind turbine foundations during low tide, the flight altitude is adjusted to plan inspection routes that are close to the wind turbine foundations and keep a safe distance from dangerous areas. Monitoring and early warning of sudden faults of wind turbines: When the wind turbine is running, vibration, temperature, speed and image data are collected in real time and transmitted to the data analysis model. The fault diagnosis model is used to deeply analyze the data. When an abnormality is identified, the wind turbine is determined to be faulty, thereby triggering an early warning, and the fault information is transmitted to the route planning system and operation and maintenance personnel; Emergency route re-planning: When a fault warning signal is received, the current tidal data is obtained to determine the tidal stage, safe flight altitude range and affected area. Based on the status of the faulty wind turbine and surrounding wind turbines, real-time tidal data, and through optimization algorithms, the impact of the tide on the surrounding area is fully considered and the existing inspection routes are comprehensively adjusted.

2. The method for autonomously planning a UAV wind turbine inspection route according to claim 1 is characterized in that: In the step of constructing the dynamic marine environment model, in combination with tidal data, a numerical simulation method is used to simulate the dynamic changes of seawater with the help of a hydrodynamic model. Specifically, the influence of the topography and coastline shape factors of the sea area on the flow of seawater is considered, and the changes in the coverage of seawater during high tide and low tide are simulated. The coverage of seawater is determined through multiple iterative calculations, and the influence of ocean dynamics principles and astronomical factors on tides is comprehensively considered to simulate the tidal flow rate and direction at different locations and times. In addition, through the analysis of historical tidal data and seabed topography data, the dangerous areas of reefs and shoals exposed after low tide are determined, and this information is integrated into the dynamic model.

3. The autonomous planning method for UAV wind turbine inspection route according to claim 1 is characterized in that: In the basic inspection route planning step, according to the dynamic marine environment model, in view of the rising sea water and the increase of floating objects at high tide, the flight altitude range is determined by comprehensively considering the tidal height, the distribution of floating objects and the flight performance of the UAV. The path planning algorithm is used to cover the cabin, blades and tower parts, and the flight distance and time cost are reduced by optimizing the path sequence and turning angles in combination with the layout and orientation of the wind turbine. At the same time, based on the distribution information of sea water and floating objects at high tide in the model, the dangerous area is set as an insurmountable obstacle to avoid it, and the route planning strategy is adjusted according to the flight data fed back by the UAV in real time.

4. The autonomous planning method for UAV wind turbine inspection route according to claim 3 is characterized in that: In the basic inspection route planning step, a path planning algorithm is used, and its algorithm formula is: Among them, C ZC is the comprehensive cost of the candidate route in the high tide stage, k is the number of key parts of wind turbines passed by the route, α, β, and γ are the weight coefficients of danger distance, flight altitude, and coverage of key parts of wind turbines, respectively. The historical inspection data and accident data are analyzed and determined using the hierarchical analysis method. is the distance between the route and the key part of the j-th wind turbine and the nearest dangerous area, d max is the maximum distance between the dangerous area and the key parts of the wind turbine under the current high tide scenario, which is determined by the dynamic offshore environment model combined with historical data and geographic information. j is the flight altitude of the UAV to reach the key part of the j-th wind turbine, which is determined by the flight altitude range set by the high tide condition, h AQ For safe flight altitude, it is determined according to tides and floating objects. j is the inspection coverage of the route for the jth key parts of the wind turbine, which is calculated based on the relationship between the preset inspection points at the key parts of the wind turbine and the route.

5. The autonomous planning method for UAV wind turbine inspection route according to claim 1 is characterized in that: In the basic inspection route planning step, the flight altitude is adjusted according to the characteristics of the wind turbine foundation being exposed at low tide. Specifically, a planning algorithm is used to plan an inspection route close to the wind turbine foundation around each inspection point of the wind turbine foundation structure in combination with its shape, size and inspection point distribution. At the same time, with the help of the dangerous area information including reefs and shoals after low tide in the model, a safety buffer zone is set around the route according to the degree of danger, the flight speed and maneuverability of the drone, and it is regarded as an inaccessible area to ensure safety. At the same time, the inspection route is adjusted according to the characteristics of the wind turbine foundation structure and important inspection parts.

6. The method for autonomously planning a UAV wind turbine inspection route according to claim 5 is characterized in that: In the basic inspection route planning step, a path planning algorithm is used to plan an inspection route close to the wind turbine foundation based on the shape, size and distribution of the inspection points of the wind turbine foundation structure. The algorithm formula is: Among them C TC is the comprehensive cost of the candidate route in the low tide stage. The lower the cost, the better the route. l is the number of wind turbine basic inspection points along the route. δ and ∈ are the weight coefficients of the distance to the dangerous area and the coverage of wind turbine basic inspection, respectively. By analyzing the historical low tide inspection data and accident data, and taking each inspection reward as feedback, δ and ∈ are adjusted to make the route adapt to environmental changes. is the distance between the route and the m wind turbine foundation inspection points and the nearest dangerous area, determined by the dynamic offshore environment model, d AQJL The safe distance is determined according to the degree of danger, which is determined by the marine environment model combined with historical data. m is the inspection detail level of the route for the mth wind turbine foundation inspection point, which is calculated based on the importance of the wind turbine foundation inspection point and the proximity between the route and the inspection point.

7. The autonomous planning method for UAV wind turbine inspection route according to claim 1 is characterized in that: In the fan sudden fault monitoring and early warning step, a fault diagnosis model is used to deeply analyze the data, and the model formula is: Among them, G is the fault judgment index. When G is greater than the threshold G max When the fan fails, it is determined that a fault has occurred and an early warning is triggered. n represents the number of data types collected, and w i is the weight coefficient of the i-th data type, which is determined by analyzing historical fault data and machine learning technology. i (x i ) is the characteristic function of the i-th data type, which is used to convert the collected raw data x i It is converted into a characteristic value that can reflect the possibility of fault. b is the weight coefficient of the comprehensive characteristics of image data, which is determined according to the importance and reliability of image data in fault judgment. g(y) is the comprehensive characteristic function of image data, and y represents the collected image data.

8. The autonomous planning method for UAV wind turbine inspection route according to claim 1 is characterized in that: In the emergency route re-planning step, the influence of tides on the surrounding areas is fully considered through the optimization algorithm, and the existing inspection routes are comprehensively adjusted. Specifically, the tide height and tide status data are retrieved in real time, and the location, type, severity of the faulty wind turbine and the real-time operating status of the surrounding wind turbines are collected. Based on these data, the fitness function is used to evaluate the candidate emergency routes, and the candidate emergency routes are generated according to the faulty wind turbine, the surrounding wind turbines and the tidal impact area. The candidate routes are genetically operated, and individuals are selected through the roulette selection method. After multiple generations of genetic operations, the route corresponding to the individual with the best fitness is selected as the emergency route. If the fitness value does not increase significantly for multiple consecutive generations, the genetic algorithm parameters are adjusted, and the emergency route is sent to the drone to perform the inspection task.

9. The method for autonomously planning a UAV wind turbine inspection route according to claim 8 is characterized in that: In the emergency route replanning step, a fitness function is used to evaluate candidate emergency routes, and the function formula is: Among them, F is the fitness value of the candidate route, which is used to evaluate the pros and cons of the candidate route. T is the time required for the UAV to reach the faulty wind turbine, which is obtained by calculating the route length and the flight speed of the UAV. The route length is calculated according to the coordinates of each point on the path. λ is the distance influence factor, which is adjusted according to the urgency of the fault and the importance of the surrounding wind turbines. D is the sum of the distances between the route and other wind turbines, which is obtained by calculating the distances between the route and the surrounding wind turbines and accumulating them. μ is the risk influence factor, which is adjusted according to the degree of danger under the current tidal conditions. R is the route risk value, which is determined by comprehensively considering the dangerous area under the influence of the tide and the dangerous range that may be caused by the faulty wind turbine.

10. The autonomous planning method for UAV wind turbine inspection route according to claim 9 is characterized in that: In the emergency route replanning step, individuals are selected by roulette selection method, and the formula is: Among them, P i represents the probability that the i-th candidate emergency route is selected for the next generation of reproduction, F i is the fitness value of the i-th candidate emergency route, N represents the population size, that is, the total number of candidate emergency routes that participate in the iterative optimization of the genetic algorithm during the emergency route replanning process, and F j is the fitness value of the jth candidate emergency route.

Images