Simulation method and system for checking wind generating set based on aircraft

Abstract

The invention discloses a simulation method and system for checking a wind generating set based on an aircraft. The simulation method comprises the following steps: acquiring meteorological data, fan design parameters and operation data and aircraft inspection task data; inputting the meteorological data into a meteorological simulation module to obtain a first meteorological simulation wind field model; inputting the first meteorological simulation wind field model and the fan design parameters and operation data into a fan dynamic simulation module to obtain a first fan dynamic simulation model; inputting the first meteorological simulation wind field model, the first fan dynamic simulation model and the aircraft inspection task data into an aircraft control simulation module to control an aircraft to execute an inspection task so as to obtain a plurality of aircraft inspection strategy sets; and inputting the plurality of aircraft inspection strategy sets into a fan defect detection module to verify the effectiveness of the strategy sets on fan defect detection, and outputting a simulation result. According to the method, an aircraft-based fan defect inspection strategy can be optimized in the aspects of efficiency, cost, safety and the like.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of wind power technology, and more particularly, to a simulation method and system for wind turbine inspection based on an aircraft. BACKGROUND

[0002] In the related art, under the impetus of the global “double carbon” goal, as one of the renewable energy sources with the most mature technology and the highest commercialization level, wind power has entered a new stage of large-scale and large-scale development. In order to improve power generation efficiency, increase wind power generation, etc., wind power technology iteration presents the trend of “turbine large-scale, blade super-long”. The single capacity of mainstream wind turbines is gradually increasing, and the length of the blades is also gradually increasing. While these changes improve power generation efficiency, they also bring unprecedented challenges to operation and maintenance management.

[0003] For example, in terms of conventional operation and maintenance difficulty, manual visual inspection is prone to missed detection, especially small cracks in key parts such as blade tips and leading edges, which are difficult to detect, resulting in safety hazards. In terms of blade design, as the length of the blade increases, the probability of failure due to insufficient blade design margin also increases because the blade design margin is compressed and the safety redundancy is insufficient. This can cause the blade to easily crack due to stress concentration in, for example, a sudden weather condition, thereby increasing the probability of wind turbine shutdown accidents due to blade cracking, which can result in significant losses. In addition, existing shutdown inspection solutions for wind turbine inspection can result in significant loss of power generation, and even further loss in the case of inspection of faults requiring shutdown for maintenance.

[0004] Therefore, how to provide a safe and effective wind turbine inspection solution is one of the problems that need to be solved at present. SUMMARY

[0005] Embodiments of the present disclosure provide a simulation method and system for wind turbine inspection based on an aircraft, which aims to solve the problem that existing wind turbine inspection solutions cannot support effective verification of wind turbine non-stop inspection solutions, and by jointly simulating the dynamic characteristics of the unit, the interaction with the complex environment, and the actual hardware control (aircraft control), a high-true, high-reliable virtual test environment is provided for actual inspection solution verification, thereby obtaining a wind turbine defect inspection strategy based on an aircraft that is further optimized in terms of efficiency, cost, safety, etc.

[0006] In one general aspect, there is provided a simulation method for aircraft-based inspection of a wind turbine generator system, the simulation method comprising: obtaining meteorological data, wind turbine design parameters and operational data, and aircraft inspection task data; inputting the meteorological data into a meteorological simulation module to obtain a first meteorological simulation wind field model; inputting the first meteorological simulation wind field model, and the wind turbine design parameters and operational data into a wind turbine dynamic simulation module to obtain a first wind turbine dynamic simulation model; inputting the first meteorological simulation wind field model, the first wind turbine dynamic simulation model, and the aircraft inspection task data into an aircraft control simulation module to control an aircraft to perform an inspection task for the wind turbine generator system to obtain a plurality of aircraft inspection strategy sets; and inputting the plurality of aircraft inspection strategy sets into a wind turbine defect detection module to verify effectiveness of the plurality of aircraft inspection strategy sets for wind turbine defect detection, and outputting a simulation result.

[0007] Optionally, before the step of obtaining meteorological data, wind turbine design parameters and operational data, and aircraft inspection task data, the simulation method can further comprise: calling and initializing a meteorological simulation module, a wind turbine dynamic simulation module, an aircraft control simulation module, and a wind turbine defect detection module, in a case where the wind turbine generator system is in an operational state.

[0008] Optionally, the step of calling and initializing the meteorological simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module, and the wind turbine defect detection module can comprise: calling a first set of operational data corresponding to the meteorological simulation module, a second set of operational data corresponding to the wind turbine dynamic simulation module, a third set of operational data corresponding to the aircraft control simulation module, and a fourth set of operational data corresponding to the wind turbine defect detection module, and combining the first set of operational data to the fourth set of operational data to obtain a fifth set of operational data, and based on the fifth set of operational data, configuring a data interface corresponding to the meteorological simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module, and the wind turbine defect detection module to enable data interaction between the modules through the corresponding data interface.

[0009] Optionally, the first wind turbine dynamic simulation model can comprise a first model based on aerodynamics and a second model based on blade element momentum theory, wherein the first model can comprise a dynamic response model based on wind speed and direction, wind turbine yawing moment, wind turbine rotational inertia, and wind turbine yawing angle, and the second model can comprise a dynamic response model based on wind speed and direction, wind turbine blade aerodynamic moment, and wind turbine blade rotational speed.

[0010] Optionally, the weather simulation module and the wind turbine dynamic simulation module can be implemented as a predetermined three-dimensional dynamic simulator, and the aircraft control simulation module can be implemented as a predetermined robot operating system terminal and a predetermined flight control terminal, wherein the predetermined robot operating system terminal and the predetermined flight control terminal cooperatively perform flight steady-state control for crew yaw dynamic identification safety monitoring based on a predetermined centering algorithm and a predetermined yaw monitoring algorithm.

[0011] Optionally, the step of inputting the weather data into the weather simulation module to obtain the first weather simulation wind field model can include: inputting real-time weather data into the weather simulation module to obtain a second weather simulation wind field model, the second weather simulation wind field model being a simulation wind field corresponding to the real-time weather data updated according to a time stamp; determining whether the second weather simulation wind field model can be effectively driven based on whether real-time wind speed and wind speed change rate obtained according to the second weather simulation wind field model meet preset conditions respectively, and taking the model determined to be able to be effectively driven as the first weather simulation wind field model.

[0012] Optionally, the step of inputting the first weather simulation wind field model and the wind turbine design parameters and operation data into the wind turbine dynamic simulation module to obtain the first wind turbine dynamic simulation model can include: inputting the first weather simulation wind field model and the wind turbine design parameters and real-time operation data into the wind turbine dynamic simulation module to obtain a second wind turbine dynamic simulation model; determining consistency between first wind turbine rotation state quantity and first wind turbine yaw state quantity acquired via a topic subscription manner and second wind turbine rotation state quantity and second wind turbine yaw state quantity obtained according to the second wind turbine dynamic simulation model by using a sliding window mean algorithm; in response to the first wind turbine rotation state quantity being consistent with the second wind turbine rotation state quantity and the first wind turbine yaw state quantity being consistent with the second wind turbine yaw state quantity, taking the second wind turbine dynamic simulation model as the first wind turbine dynamic simulation model.

[0013] Optionally, before controlling the aircraft to perform the inspection task for the wind turbine generator, the simulation method can further include: determining whether the aircraft can remain at a target position under a specific climate condition (for example, a specific weather condition) and a specific wind turbine operation state based on real-time acquired aircraft power state, the first weather simulation wind field model and the first wind turbine dynamic simulation model; in response to the aircraft being able to remain at the target position under the specific climate condition (for example, the specific weather condition) and the specific wind turbine operation state, calibrating each component of the aircraft and each control parameter to control the aircraft to perform the inspection task after calibration.

[0014] In another general aspect, there is provided a simulation system for aircraft-based wind turbine inspection, the simulation system comprising: a meteorological simulation module; a wind turbine dynamic simulation module; an aircraft control simulation module; a wind turbine defect detection module; and a controller configured to implement the simulation method for aircraft-based wind turbine inspection as described above by utilizing the meteorological simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module, and the wind turbine defect detection module.

[0015] In another general aspect, there is provided a computer-readable storage medium that, when instructions in the computer-readable storage medium are executed by at least one processor, cause the at least one processor to perform the simulation method for aircraft-based wind turbine inspection as described above.

[0016] In another general aspect, there is provided a computer program product comprising computer programs / instructions that, when executed by a processor, implement the simulation method for aircraft-based wind turbine inspection as described above.

[0017] In another general aspect, there is provided a computer device comprising: at least one processor; at least one memory storing computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform the simulation method for aircraft-based wind turbine inspection as described above.

[0018] The simulation method and system for aircraft-based wind turbine inspection according to embodiments of the present disclosure provide a high-fidelity and high-reliability virtual test environment for actual inspection scheme verification by jointly simulating the dynamic characteristics of the turbine, the complex environment interaction, and the actual hardware control (e.g., aircraft control), thereby enabling an aircraft-based wind turbine defect inspection strategy that is further optimized in terms of efficiency, cost, safety, and the like, such as accelerating the verification and final determination of the aircraft-based wind turbine defect inspection scheme, improving the safety and reliability of the inspection process, reducing the development cost of the scheme, and the like. In addition, the simulation method for aircraft-based wind turbine inspection according to embodiments of the present disclosure breaks through the limitations of existing general simulation tools in terms of adaptability to wind power scenarios, implements deep dynamic coupling of the meteorological environment, the turbine operating state, and the aircraft, and effectively solves the problem of static simulation environment and disconnection from reality according to the real-time environment simulation scheme, thereby providing a highly realistic real-world scenario for aircraft inspection simulation. BRIEF DESCRIPTION OF DRAWINGS

[0019] The above and other objects and features of embodiments of the present disclosure will become more apparent from the following description made with reference to the accompanying drawings, in which: FIG. 1 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to an embodiment of the present disclosure; FIG. 2 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 3 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 4 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 5 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 6 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 7 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 8 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 9 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 10 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 11 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 12 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 13 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 14 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 15 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 16 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 17 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 18 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 19 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 20 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 21 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 22 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 23 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 24 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 25 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 26 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 27 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 28 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 29 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 30 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 31 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 32 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 33 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 34 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 35 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 36 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 37 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 38 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 39 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 40 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 41 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 42 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 43 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 44 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 45 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 46 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 47 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 48 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 49 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 50 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 51 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 52 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 53 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 54 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present disclosure; FIG. 55 shows a schematic diagram of a simulation system for aircraft-based wind turbine inspection according to another embodiment of the present Figure 1 is a flow chart showing a simulation method for aircraft-based wind turbine generator set inspection according to an embodiment of the present disclosure; Figure 2 is a block diagram showing a simulation system for aircraft-based wind turbine generator set inspection according to an embodiment of the present disclosure; Figure 3 is a flow chart showing an example of a simulation method for aircraft-based wind turbine generator set inspection according to an embodiment of the present disclosure; Figures 4 to 12 is a simulation process schematic diagram and a simulation effect diagram of a simulation method for aircraft-based wind turbine generator set inspection according to an embodiment of the present disclosure; and Figure 13 is a block diagram showing a computer device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0020] The following detailed description is provided to aid the reader in understanding the methods, devices, and / or systems described herein. However, various changes, modifications, and equivalents of the methods, devices, and / or systems described herein will be clear to those skilled in the art after understanding the present disclosure. For example, the order of the operations described herein is merely an example, and is not limited to those set forth herein, but can be changed as will be clear after understanding the present disclosure, except for operations that must occur in a specific order. In addition, the description of features known in the art can be omitted for the sake of more clarity and conciseness.

[0021] Reference will now be made in detail to embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like elements throughout. The embodiments will be explained by referring to the drawings in order to explain the present disclosure.

[0022] The single capacity of a wind turbine (referred to as a wind turbine) has been upgraded from 3-4 MW to 5-8 MW, and the offshore wind turbine even exceeds 15 MW; the length of the matching blade has been increased from 40-50 m to 70-100 m, and the length of the blade of some offshore wind turbines has exceeded 120 m, the weight of a single blade can reach more than 50 tons, the height of the wind turbine tower is generally more than 100 m, and the highest is more than 160 m, and the wind power plant is transforming from a "small and medium-sized wind turbine cluster" to a "super large-scale layout of wind turbine".

[0023] However, the large-scale of the unit and the super-long of the blade bring challenges to operation and maintenance while improving power generation efficiency. For example, in one aspect, the traditional manual inspection requires the staff to climb the 100-meter tower drum and face the inspection of the super-long blade (for example, the 100-meter blade needs to be inspected in sections), and the single inspection time may be extended from 2 to 4 hours to 6 to 8 hours, causing the labor intensity to double, and the super-long blade inspection needs to cover a larger range (for example, the surface area of the blade is more than 2000m 2 ), and manual visual detection is prone to missed detection, especially the tiny cracks in the key parts such as the blade tip and leading edge, which are difficult to be found, thereby hiding the safety hazard, so in this case, the operation and maintenance of the wind turbine / unit are more difficult.

[0024] For example, in another aspect, in the above case, in order to balance the power generation efficiency and the material cost, the super-long blade usually adopts lightweight and high specific strength composite materials (for example, carbon fiber reinforced resin matrix composite materials), and the design safety factor is compressed from the traditional 1.5-2.0 to 1.2-1.3, and the ability to resist extreme working conditions is significantly reduced. For example, in the case of strong wind, thunderstorm, salt spray (offshore), low temperature freezing and other sudden conditions, the blade is prone to crack propagation due to stress concentration. For another example, as the length of the blade increases, the probability of failure due to insufficient design margin may increase, thereby causing serious loss of power generation. Therefore, in this case, the design margin of the blade is compressed, and the safety redundancy may be insufficient.

[0025] For example, in yet another aspect, in the above case, the cost of shutdown inspection is high: the daily average power generation of a super-large wind turbine is tens of thousands of degrees, and the traditional shutdown inspection (for example, 4 to 8 hours each time) may lose tens of thousands of degrees or more; in the case of long-term shutdown for maintenance (average maintenance period of 7 to 15 days) due to blade failure, the power generation loss of a single wind turbine is more than 200,000 degrees, thereby seriously affecting the efficiency of the wind farm.

[0026] In view of the above and other technical problems existing in the prior art, the present disclosure proposes a simulation method and system for wind turbine inspection based on an aircraft, which can at least realize: by constructing a dynamic joint simulation system of a wind turbine and an aircraft (for example, a drone) and implementing the corresponding simulation method, the problem that the existing wind turbine inspection simulation technology cannot meet the operation and maintenance scheme simulation of the large-scale of the unit, the super-long of the blade and the loss of power generation due to shutdown can be solved.

[0027] For example, by reproducing the complex physical field environment around the fan (for example, an ultra-large fan), various different meteorological coupling effects are fully considered, so that the simulation data is close to the actual collected data, and the effectiveness verification of the related algorithm is effectively supported. By reproducing the actual operation condition of the fan, the unmanned aerial vehicle simulation data can be adjusted in real time for the yaw, rotation and other actions of the fan, so that the unmanned aerial vehicle inspection control strategy is more in line with the real fan operation state, thereby improving the accuracy, safety and reliability of the inspection control strategy. By realizing the unmanned aerial vehicle inspection control strategy simulation under the condition that the wind turbine generator set does not stop, a high-accuracy, high-reliability virtual test environment is provided for actual inspection scheme verification.

[0028] The simulation method and system for wind turbine generator set inspection based on an aerial vehicle according to embodiments of the present disclosure are described below with reference to the accompanying drawings. Figures 1 to 13 The simulation method and system for wind turbine generator set inspection based on an aerial vehicle according to embodiments of the present disclosure are described below with reference to the accompanying drawings.

[0029] First, referring to Figure 1 and Figure 3 The simulation method and system for wind turbine generator set inspection based on an aerial vehicle according to embodiments of the present disclosure are described below with reference to the accompanying drawings.

[0030] Figure 1 is a flowchart illustrating a simulation method for wind turbine generator set inspection based on an aerial vehicle according to embodiments of the present disclosure. Figure 3 is a flowchart illustrating an example of a simulation method for wind turbine generator set inspection based on an aerial vehicle according to embodiments of the present disclosure.

[0031] Referring to Figure 1 , according to embodiments of the present disclosure, in step S101, meteorological data, fan design parameters and operation data and aerial vehicle inspection task data are acquired. For example, the meteorological data, fan design parameters and operation data and aerial vehicle inspection task data can be acquired from

[0032] Before step S101, the simulation method 100 can further include the following steps: in the case that the wind turbine generator set is in an operation state (which means that the entire simulation system is executed when the simulation method is executed, without stopping the wind turbine generator set but in an operation state), calling and initializing the meteorological simulation module, the fan dynamic simulation module, the aerial vehicle control simulation module and the fan defect detection module.

[0033] For example, in the case that the wind turbine is in the running state, the weather simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module and the wind turbine defect detection module can be called and initialized in sequence. Here, the weather simulation module and the wind turbine dynamic simulation module can be implemented as a predetermined three-dimensional dynamic simulator (for example, Gazebo, Unity, AirSim). In addition, the aircraft control simulation module can be implemented as a predetermined robot operating system terminal (for example, ROS) and a predetermined flight control terminal (for example, PX4 firmware).

[0034] As an example, each of the above modules can be called by a controller of a predetermined operating system as a simulation system. For example, the Ubuntu operating system is used as the simulation system to provide underlying environment support, and the integrated simulation main process is triggered to start by executing the launch / main.launch main script.

[0035] By sequentially calling each of the above modules, it is beneficial to subsequently implement data communication of each of the above modules to interconnect, and it is beneficial to subsequently derive an optimal inspection strategy that comprehensively considers each aspect embodied by each of the above modules.

[0036] Further, as an example, the steps of calling and initializing the weather simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module and the wind turbine defect detection module can specifically include steps S11 and S12: In step S11, a first running data set corresponding to the weather simulation module, a second running data set corresponding to the wind turbine dynamic simulation module, a third running data set corresponding to the aircraft control simulation module and a fourth running data set corresponding to the wind turbine defect detection module are called, and the first running data set to the fourth running data set are combined (for example, sequentially combined) to obtain a fifth running data set.

[0037] For example, the above simulation main process can be used to implement the initialization scripts of the wind turbine dynamic simulation module (described later using Gazebo as an example), the unmanned aerial vehicle control simulation module (described later using ROS as an example), the weather simulation module and the wind turbine defect identification module in series. The module loading order is defined by an XML configuration file, for example, the weather simulation module, the wind turbine dynamic simulation module, the unmanned aerial vehicle control simulation module and the wind turbine defect identification module are loaded in sequence.

[0038] In step S12, based on the fifth running data set, the data interfaces corresponding to the weather simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module and the wind turbine defect detection module are configured for communication, so that the modules can realize data interaction through the corresponding data interfaces.

[0039] For example, the initialization of the ROS distributed communication node can be realized by the simulation master process described above, the master communication node (sim_master) is created by, for example, the roscpp library, and the topics (Topics) and services (Services) of each sub-module are uniformly managed; the bridge configuration of ROS and Gazebo is completed, the real-time interaction between the physical simulation data (for example, the pose of the unmanned aerial vehicle, the rotating speed of the fan, etc.) and the ROS node is realized through gazebo_ros; the PX4 firmware is compiled, the simulation control of the quadrotor unmanned aerial vehicle (including the motor control interface, sensor data analysis logic, etc.) is generated, and the firmware is confirmed to be loaded normally through the initialization self-check, and finally the basic conditions for multi-module integrated simulation are met.

[0040] By enabling each of the above modules to realize data interaction with each other, a simulation system suitable for fan non-stop inspection is constructed. In this way, the limitations of existing general simulation tools in the adaptability of wind power scene are broken through, and deep dynamic coupling of the "environment-fan running state-aircraft" is realized. This not only reflects the motion interaction in the physical layer, but also involves the coordination of control logic, data communication and other aspects, breaking through the limitations of traditional isolated model simulation, and fully simulating the dynamic process of the fan and the aircraft in actual operation.

[0041] According to an embodiment of the present disclosure, in step S102, meteorological data is input into a meteorological simulation module to obtain a first meteorological simulation wind field model.

[0042] As an example, step S102 can further include step S1021 and step S1022: In step S1021, real-time meteorological data is input into the meteorological simulation module to obtain a second meteorological simulation wind field model.

[0043] Here, the second meteorological simulation wind field model is a simulation wind field corresponding to the real-time meteorological data updated according to the timestamp.

[0044] For example, the following preprocessing can be performed on the real-time meteorological data before inputting it into the meteorological simulation module: taking the real-time monitoring data of the ground meteorological station (such as wind speed, wind direction, temperature, etc.), the output data of the numerical weather prediction model (WRF) as the data source, filtering the netCDF4 format data according to the simulation area and time granularity (such as 10-minute time granularity), and analyzing the data structure through the xarray Python library, converting the meteorological data into a Gazebo recognizable XML format text (including wind_speed, wind_direction, turbulence_intensity, etc. Data tags).

[0045] At step S1022, whether the second meteorological simulation wind field model can be effectively driven is determined based on whether the real-time wind speed and the wind speed change rate obtained according to the second meteorological simulation wind field model meet preset conditions respectively, and the model determined to be able to be effectively driven is taken as the first meteorological simulation wind field model.

[0046] For example, the above-mentioned wind field simulation and driving verification operation can be implemented by the following steps: calling the Gazebo built-in wind_plugin plug-in to read the converted data, referencing and dynamically updating the wind field parameters (such as wind parameters refreshed every 10 seconds) in the simulation world file (windmill.world) to achieve high-fidelity simulation of the actual wind field. In addition, the meteorological data driving effectiveness verification can also be performed for the key link: by comparing the real-time wind speed output by the above-mentioned plug-in with the preset threshold (such as the fluctuation range of ±0.5 m / s under the rated wind speed of 12 m / s), and combining the wind speed change rate (≤2 m / s²) to judge whether the driving logic is normal; if the verification fails, the main process triggers the data reloading mechanism (reloads the meteorological data of the last 10 minutes, for example).

[0047] By constructing a “terrain-meteorology-airflow” three-dimensional coupled simulation model, at least the following can be achieved: by adapting to complex terrains such as mountainous areas (moderate slopes) and sea areas (considering salt spray), the influence of terrain on local wind direction is simulated; by connecting real-time meteorological interfaces, various parameters such as wind speed, wind direction, temperature, and humidity are dynamically introduced, and complex coupled effects such as “high temperature and high humidity” and “low temperature and strong wind” are simulated synchronously; by constructing airflow models such as tower cylinder wake and blade vortex ring, the interference of airflow on the flight attitude of the aircraft is accurately restored, thereby ensuring that the simulation environment is highly consistent with the actual operation and maintenance scene; further covering key meteorological variables such as super-large wind turbine operation, the operation environment simulation of super-large wind turbines is more accurate, for example, by simulating the influence of complex airflow (such as tower cylinder wake and blade vortex ring) around the super-large wind turbine on the flight attitude of the aircraft, a corresponding flight path planning scheme is simulated and verified.

[0048] Through the judgment of the effectively driven model, it is ensured that the multiple physical driving influence fields of “environment-wind turbine-aircraft” are accurately effective, and the subsequent simulation deviation caused by abnormal wind field data is avoided. In addition, by enabling the wind turbine and the aircraft to function in the same meteorological condition field, the important differences of multi-system comprehensive simulation of meteorology-wind turbine, meteorology-aircraft, and aircraft-sensor-wind turbine can also be reflected.

[0049] According to an embodiment of the present disclosure, at step S103, the first meteorological simulation wind field model and the wind turbine design parameters and operation data are input into a wind turbine dynamic simulation module to obtain a first wind turbine dynamic simulation model.

[0050] As an example, the first wind turbine dynamic simulation model can include a first model based on aerodynamics and a second model based on blade element momentum theory. Here, the first model can include a dynamic response model based on wind speed and wind direction, wind turbine yawing moment, wind turbine rotational inertia, and wind turbine yawing angle, and the second model can include a dynamic response model based on wind speed and wind direction, wind turbine blade aerodynamic moment, wind turbine blade rotational speed.

[0051] In particular, as an example, in the case of a wind turbine being an offshore wind turbine (for example, an offshore wind turbine with a capacity of 6 MW, a tower height of 130 m, and a blade length of 90 m), a wind turbine dynamic simulation model including the following models can be constructed in Gazebo through a Simulation Description Format (SDF) file: Wind speed and wind direction v-yawing moment M-rotational inertia J-yawing angle Model: based on aerodynamic theory, establish yawing moment calculation formula M=C y (α) 1 / 2ρv vS L (where C y represents the yawing force coefficient, α represents the wind direction angle, ρ represents the air density, S represents the cabin windward area of the wind turbine, and L represents the yawing moment arm); combined with the cabin rotational inertia J=∫r rdm (for example, about 5×10 kg m 2 ), through the differential equation M=J Establish the dynamic response relationship of the yawing angle with the wind speed v.

[0052] As an example of the second model, the wind speed and wind direction v-aerodynamic moment T-rotational speed ω model: based on blade element momentum theory, divide the blade into a predetermined number of units (for example, 50 units), calculate the aerodynamic moment of each unit and integrate to obtain the total aerodynamic moment T; combined with the transmission system rotational inertia Jt=2×10000 kg m 2 , through T=Jt Establish the transfer function of rotational speed ω and wind speed v (for example, wind speed 8 m / s corresponds to rotational speed 10 rpm, and wind speed 12 m / s corresponds to rotational speed 18 rpm).

[0053] By constructing a fan dynamic simulation model including a first model and a second model, the dynamic response of the fan under various meteorological conditions can be accurately reflected. By deeply integrating "blade rotation-yaw control-power output-torque control" in the simulation, based on aerodynamics, mechanical dynamics, etc., the mathematical relationship between wind speed and various operating parameters of the fan is accurately established, thereby truly reproducing the dynamic running process of the fan in the non-stop state, and realizing the interaction between the multi-dimensional fan dynamic model and the real-time environment. The interaction of this real-time environment simulation technology effectively solves the problem of static and disconnection with reality of the traditional simulation environment, and provides a highly realistic real scene for subsequent aircraft inspection (such as unmanned aerial vehicle inspection) simulation. Further, by combining meteorological simulation and fan dynamic simulation to accurately predict unmanned aerial vehicle inspection risks, such as early warning of the impact of severe weather on inspection, it is helpful to generate a scientific and reasonable inspection plan for the unmanned aerial vehicle.

[0054] As an example, step S103 can further include steps S1031 to S1033: In step S1031, the first meteorological simulation wind field model and the fan design parameters and real-time operating data (including real-time operating data of driving data) are input into the fan dynamic simulation module to obtain a second fan dynamic simulation model.

[0055] In step S1032, a sliding window mean algorithm is used to determine the consistency between the first fan rotation state quantity and the first fan yaw state quantity obtained through topic subscription and the second fan rotation state quantity and the second fan yaw state quantity obtained according to the second fan dynamic simulation model.

[0056] For example, by subscribing to the fan rotation state quantity (R_st, or rotation signal, including blade speed, transmission torque) and the yaw state quantity (Y_st, or yaw signal, including yaw angle, yaw speed) through the ROS topic / turbine / status (its specific meaning is that the two ROS topics with the names turbine and status), the sliding window mean algorithm (for example, the window size is 50 ms) is used to verify the consistency of the signal and the output of the above fan mathematical model, that is, the error threshold control of maximum, minimum, mean, variance of the time series data in the sliding window content. For example, when the speed error is greater than 2% or the yaw angle error is greater than 1°, the parameter correction of the control submodule is triggered (for example, adjusting the torque output coefficient Kt to 0.85). Therefore, through the above R / Y signal consistency verification, it can be ensured that the fan model can accurately respond to the meteorological wind field and provide a real dynamic target for unmanned aerial vehicle inspection.

[0057] Here, regarding the above ROS topic / turbine / status, the following is described: in ROS (Robot Operating System), a communication mechanism is for different nodes (i.e., Node, which can be a program module running independently, for example) to exchange data, work cooperatively, and Topic Subscription is an asynchronous communication mode in ROS, and is implemented based on a "Publish-Subscribe" model. A Topic is a "channel" for passing data of a specific type between nodes, and its communication process includes a Publisher and a Subscriber. For example, a Publisher can actively send data to a specific Topic (such as a sensor node publishing camera image, laser radar point cloud), and a Subscriber can passively receive all data on the Topic by "subscribing" to the specific Topic (such as a control node subscribing to sensor data for decision-making). Both the Publisher and the Subscriber are associated through the same Topic name, and there is no need to know the existence of the other party in advance (i.e., decoupled design), the Publisher continuously sends data, and the Subscriber receives data on demand, thereby being suitable for high-frequency, continuous data transmission (such as real-time sensor data, wind turbine state, UAV pose, etc.).

[0058] In step S1033, in response to the first wind turbine rotation state quantity and the second wind turbine rotation state quantity being consistent and the first wind turbine yaw state quantity and the second wind turbine yaw state quantity being consistent, the second wind turbine dynamic simulation model is taken as the first wind turbine dynamic simulation model.

[0059] The first wind turbine dynamic simulation model obtained through the above steps can ensure that the wind turbine simulation model can accurately respond to the meteorological wind field to perform corresponding actions, and provide a real dynamic target for aircraft inspection. For example, dynamic modeling of a wind turbine generator set and signal checking for yaw signals and rotation signals are achieved.

[0060] According to an embodiment of the present disclosure, in step S104, the first meteorological simulation wind field model, the first wind turbine dynamic simulation model, and the aircraft inspection task data are input into an aircraft control simulation module to control the aircraft to perform an inspection task for the wind turbine generator set, to obtain a plurality of aircraft inspection strategy sets.

[0061] For example, each aircraft inspection strategy set in the plurality of aircraft inspection strategy sets can include simulation appearance troubleshooting data obtained by the aircraft when performing the inspection task (for example, wind turbine blade image data taken by the aircraft when performing the inspection task) and a UAV trajectory (for example, flight attitude and route data when checking). For example, each inspection strategy set can further include wind turbine operation logs, UAV logs, and the like associated with the inspection task.

[0062] As an example, by combining the first meteorological simulation wind field model and the first wind turbine dynamic simulation model to simulate instantaneous instability scenarios, a realistic dynamic input is provided for aircraft obstacle avoidance and trajectory correction.

[0063] Here, as an example, before controlling the aircraft to perform an inspection task on the wind turbine generator, the simulation method 100 may also include steps S41 and S42: In step S41, based on the real-time acquired aircraft power status, the first meteorological simulation wind field model, and the first wind turbine dynamic simulation model, it is determined whether the aircraft can maintain the target position under specific climatic conditions (e.g., specific meteorological conditions) and specific wind turbine operating conditions.

[0064] Specifically, firstly, takeoff control at a reference altitude can be achieved. For example, based on the altitude closed-loop control algorithm of the flight control firmware PX4, the output power of the multi-rotor motor (taking a quadcopter motor as an example) is adjusted (e.g., the initial power of the first motor accounts for 40%) to drive the UAV to take off vertically from the takeoff and landing point to a predetermined altitude (e.g., 1.5m). The attitude estimation is optimized by fusing IMU (Inertial Measurement Unit) and GPS data (e.g., Extended Kalman Filter (EKF)) to control the hovering error to less than or equal to the preset reference altitude (e.g., 0.1m). This reference altitude is the altitude corresponding to the reference position of the UAV when performing self-checks and mission preparation.

[0065] Then, a specific self-test under wind conditions can be simulated based on a power-location algorithm. Specifically, this can be based on a drone battery model (e.g., in the case of a 22000mAh capacity, the battery discharge curve fitting formula is: U=14.8). 0.002Q, where Q represents discharge capacity) to monitor the power status in real time; combined with, for example, heuristic search ( The path planning algorithm verifies the position holding capability under wind speed interference of, for example, 5 m / s (e.g., setting the target position deviation to be less than or equal to 0.3 m), while triggering the battery warning logic (i.e., publishing the / drone / battery_warn topic when the remaining battery power is 20%).

[0066] Next, sensor and motor algorithm verification can be performed, for example, for the laser radar (variable Lidar_st): through a predetermined algorithm, the coincidence degree of adjacent frame rate point clouds is compared (for example, the value is required to be greater than or equal to 95%), and the data accuracy is verified; for the camera (variable Cam_st): through the SIFT feature extraction algorithm, the number of feature points of the standard target (such as a checkerboard) is detected (for example, the number is required to be greater than or equal to 50), and the image validity is verified; for the quadcopter motor (variables Motor1-4_st): a PID speed closed-loop control is used (for example, the target speed is 1500 rpm, the proportional coefficient Kp is 0.2, the integral coefficient Ki is 0.05, and the differential coefficient Kd is 0.01), and the speed error is controlled to be less than or equal to 50 rpm, and the driving logic is verified.

[0067] At step S42, in response to the fact that the aerial vehicle is able to maintain the target position under the specific weather condition (e.g., specific meteorological condition) and the specific fan operation state, each component of the aerial vehicle and each control parameter are calibrated to control the aerial vehicle to perform the inspection task after calibration.

[0068] Specifically, the following full signal consistency verification can be performed: using the TF2 clock synchronization protocol (time synchronization error less than or equal to 10 ms) in ROS, the variables Lidar_st, Cam_st, Motor1-4_st and the unmanned aerial vehicle pose state variable (Drone_st, including NED coordinates, roll / pitch / yaw angles) are cross-module consistency verified. For example, by comparing the time stamps and data formats of each state variable (for example, the data format of the laser radar point cloud is PCD format, and the data format of the camera image is JPEG format), it is ensured that there is no data misplacement or loss; after the above consistency verification passes, the main process publishes the / drone / self_check_pass topic, allowing entering the inspection task phase.

[0069] The power-position algorithm self-check and sensor and motor algorithm verification of the aerial vehicle in combination with the meteorological wind field and the dynamic response of the fan can enable the aerial vehicle to perform parameter calibration considering the environment and the fan operation state, thereby making full and sufficient preparations for performing the inspection task.

[0070] In addition, for example, the predetermined robot operating system terminal and the predetermined flight control terminal cooperatively perform flight steady state control for crew yaw dynamic identification safety monitoring based on a predetermined centering algorithm and a predetermined yaw monitoring algorithm.

[0071] As an example, under the continuous driving of the meteorological wind field, the fan performs dynamic actions according to the above-mentioned fan dynamic simulation model (for example, the corresponding blade rotating speed is 12 rpm when the wind speed is 8 m / s, and the yaw action is triggered when the wind direction changes by 20°), and the unmanned aerial vehicle takes off from a reference height of, for example, 1.5 m. In the scenario where a double laser radar (L1, L2) is arranged on the unmanned aerial vehicle, a predetermined centering algorithm can be implemented as follows: the laser radar L1 (installed at the front end of the unmanned aerial vehicle) scans the hub center of the fan, extracts the three-dimensional coordinates of the hub (for example, N = 100 m, E = 20 m, D = -80 m based on the NED coordinate system) through a point cloud clustering algorithm, calculates the relative position deviation of the unmanned aerial vehicle and the hub (for example, ΔX = 5 m, ΔY = 8 m), and drives the unmanned aerial vehicle to adjust the roll angle (for example, the value is ≤3°) and the pitch angle (for example, the value is ≤3°), and finally stops at a predetermined distance (for example, 30 m) in front of the hub (for example, the centering error is ≤0.3 m).

[0072] Here, the NED coordinate system is a local geographic coordinate system, which is widely used in navigation and inertial measurement. The NED coordinate system includes north, east and down (pointing to the center of the earth), which belongs to the right-handed coordinate system. The origin is usually selected at the current position of the target object, and the coordinate plane is the WGS 84 ellipsoid tangent plane. Since the axial direction directly corresponds to the natural orientation, NED can express the attitude, speed, position and other states of, for example, an aircraft in a three-dimensional space in an intuitive and unified manner, and facilitate conversion with global coordinate systems (for example, ECEF, WGS 84).

[0073] In the case of the above-mentioned example, a predetermined yaw monitoring algorithm can be implemented as follows: the laser radar L2 (installed at the side end of the unmanned aerial vehicle) scans the cabin yaw angle in real time, and generates a yaw angle time series curve (for example, the data sampling frequency is 10 Hz); when the monitored yaw angle change is greater than 5°, the task point of the unmanned aerial vehicle is dynamically updated (for example, when the yaw is 10°, the task point is translated by 2 m along the yaw direction), so as to ensure that the inspection target always covers the key areas such as the blade leading edge and the blade tip.

[0074] Further, in the case of the above example, the safety monitoring and PID steady-state control of the unmanned aerial vehicle can be achieved in the following manner: a machine set yaw dynamic identification safety monitoring module based on dual laser radar (L1, L2) fusion data is started, and the yaw angle change rate of the wind turbine (for example, its value is ≤0.5° / s) and the minimum safety distance between the unmanned aerial vehicle and the blade (for example, its value is ≥3m) are calculated in real time; the following point-to-point flight steady-state control is achieved using an incremental PID control algorithm: the input of the flight steady-state control is the target waypoint coordinates (for example, N=100m, E=50m, D=-80m); the output of the flight steady-state control is the roll angle adjustment amount Δγ, the pitch angle adjustment amount Δθ, and the flight speed v (for example, the horizontal speed is 2m / s, and the vertical speed is 1m / s); the control target of the flight steady-state control is the Y variable (i.e., the safety distance between the unmanned aerial vehicle and the blade), and the error is controlled to be less than or equal to 0.5m, and the yaw identification angle and the actual yaw angle of the wind turbine are controlled to deviate by less than or equal to 1°; through PID parameter self-adaptive adjustment (for example, when the wind speed is >8m / s, Kp is increased to 0.3), the unmanned aerial vehicle can maintain a safe and stable inspection posture during the dynamic operation of the wind turbine, and the risk of collision caused by the yaw of the wind turbine or the rotation of the blade is avoided.

[0075] As another example, more accurate spatial perception and obstacle avoidance can be achieved through the technology of mapping camera photos to space. Specifically, the camera carried by the aerial vehicle is calibrated to establish a mapping relationship between image pixels and actual spatial coordinates, and the distance and relative position relationship between the aerial vehicle and the blade in space are calculated to determine whether there is a collision risk. If a potential collision is detected, the aerial vehicle control system will plan an obstacle avoidance path according to the spatial information, thereby achieving autonomous obstacle avoidance.

[0076] By using the control logic (i.e., the predetermined centering algorithm and the predetermined yaw monitoring algorithm (for example, the blade obstacle avoidance algorithm and the dynamic correction algorithm)) of the predetermined flight control terminal for the wind power inspection scene, and by means of the multi-source data fusion technology of the laser radar, visual sensor, etc. on the aerial vehicle, through complex target recognition and path planning algorithms, the actual aerial vehicle control is highly simulated, thereby achieving intelligent obstacle avoidance of the aerial vehicle when approaching the wind turbine blade. In addition, the parameters of the laser radar and the camera carried by the aerial vehicle can be kept consistent with the actual situation, and the effectiveness of the simulation parameters is strong.

[0077] According to an embodiment of the present disclosure, in step S105, the plurality of aerial vehicle inspection strategy sets are input to the wind turbine defect detection module to verify the effectiveness of the plurality of aerial vehicle inspection strategy sets for wind turbine defect detection, and a simulation result is output.

[0078] As an example, through the fan defect detection module, complex defects such as irregular cracks, point corrosion spots, etc. can be simulated and used for subsequent defect identification; interference factors such as light and blade deformation can be simulated, so that the defect identification accuracy greatly reduces the actual deviation; and the defect identification result can be fed back to the aircraft to optimize the linkage logic of the aircraft hovering, zooming, etc., so as to realize the whole process of "identification-decision-control" effectively.

[0079] By deeply integrating key processes such as "weather-driven verification-fan-driven verification-aircraft multi-dimensional self-check-PID steady-state control-photographing timing linkage-homing planning" with robot bottom technologies (for example, ROS, Gazebo, PX4), the dynamic integration true process of the wind turbine generator set under non-stop state and the aircraft is reproduced, a high-precision, quantifiable virtual test environment for safety and effectiveness verification of the actual inspection scheme is provided, and finally a simulation result indicating a strategy set with the highest effectiveness in a plurality of aircraft inspection strategy sets is obtained.

[0080] In addition, the simulation method and system of the present disclosure do not need to be implemented for specific models, have strong universality, low adaptation cost, are suitable for various types of fans, for example, have remarkable effect in application to inspection of super-large units, and can meet the rapid iteration operation and maintenance needs of wind farms.

[0081] Here, an example of a simulation method for wind turbine generator set inspection based on an aircraft according to an embodiment of the present disclosure is described with reference to Figure 3

[0082] With reference to Figure 3 In step S301, the main process of the simulation method is started. For example, the weather simulation module, the fan dynamic simulation module, the aircraft control simulation module and the fan defect detection module are called and initialized.

[0083] In step S302, it is judged whether the weather data (for example, weather data obtained from netCDF4) driving is normal.

[0084] In step S303, a fan dynamic mathematical model is constructed, for example, a wind speed and wind direction v-aerodynamic moment T-rotational speed ω mathematical model is established, a wind speed and wind direction v-yaw moment M-rotational inertia J-yaw angle Mathematical model is established, and the R quantity (R signal) and the Y quantity (Y signal) of the fan are driven.

[0085] In step S304, it is verified whether the R / Y signal of the fan is normal.

[0086] ​At step S305, in response to the R / Y signal of the fan being normal, the corresponding indicators and variables of each component of the UAV in the simulated wind field are checked. For example, in the case of the UAV taking off 1.5 m, the power position algorithm self-check Drone_st, the laser radar sensor data algorithm check Cam_st, the four-rotor motor Motor(1-4)_st, and the full signal quantity consistency algorithm check are performed in the simulated wind field.

[0087] At step S306, it is verified whether all state signals are normal.

[0088] At step S307, in response to all state signals being normal, the fan is driven to rotate and yaw in the meteorological wind field, and the UAV takes off in the same meteorological wind field and performs the predetermined centering algorithm and the predetermined yaw detection algorithm according to the data of the laser radars L1 and L2, so as to constantly update the task point position.

[0089] At step S308, the yaw dynamic recognition safety monitoring is performed according to the laser radars L1 and L2, and the point-to-point flight steady-state control is realized through the error control of the PID.

[0090] At step S309, it is judged whether the yaw recognition angle meets the predetermined Y variable error.

[0091] At step S310, in response to the yaw recognition angle meeting the predetermined Y variable error, the fan blades are photographed at the task point position.

[0092] Specifically, at each preset task point position (for example, taking the 93 m blade as an example, 30 m before and after the blade horizontal plane, horizontal left and right moving step 25 m, vertical up and down moving step 30 m), the UAV control system performs data acquisition and control according to the following logic: first, real-time acquisition of key parameters, acquisition of real-time rotating speed of the blade through the laser radar L1 (for example, 15 rpm), recording of the rotating speed fluctuation period T_0=4s (corresponding to the rotating speed 15 rpm); acquisition of the blade surface image through the camera, and synchronous recording of the time delay parameters of the two items: laser radar data transmission time delay T_e0 (calculated by the difference between the data publishing time and the receiving time of the laser radar L1, for example, the value is ≤20 ms), camera shooting execution time delay T_e1 (calculated by the difference between the trigger instruction sending time and the image output time, for example, the value is ≤30 ms). Secondly, a photographing timing signal is generated, the rotating blade rotating speed recognition photographing interval T is calculated, and the photographing timing signal is output through the ROS service / camera / trigger: after the T_e0 time delay ends, it is ensured that the laser radar has acquired the latest rotating speed data, and the camera is triggered to enter the preparation state; after delaying for the time delay T_e1, the photographing instruction is sent, and the blade image screenshot is completed (for example, the image resolution is 1920×1080, and the image definition ratio is ≥90%), so as to avoid image blur caused by time delay.

[0093] In step S311, it is determined whether the current task point is the last task point.

[0094] If the current task point is not the last task point, the drone is triggered to fly to the next point. For example, after taking a picture, the coordinates of the next task point are published via rostopic / drone / next_waypoint, driving the drone to fly from the current position to the new point according to the PID control logic described above, thus realizing the continuous execution of the inspection task.

[0095] In step S312, in response to the current task point being the last task point, the return-to-home planning algorithm is executed based on the UAV state variables and crew yaw, and the return-to-home operation is performed.

[0096] For example, the following return-to-base plan is executed: after each inspection of a task point is completed, the control system determines whether the current task point is the last task point by querying the task list (for example, there are twelve task points in the case of three blades and four points).

[0097] If this is the last task point, then based on the UAV's current pose (e.g., latitude and longitude N30°00′, E120°00′) and the coordinates of the take-off and landing points (e.g., latitude and longitude N30°03′, E120°02′), a preset path generation algorithm (e.g., but not limited to, heuristic search-based algorithm) is invoked. The system employs three return-to-home planning algorithms: one based on rapid-exploring random trees (RRT), and another based on model predictive control (MPC). These algorithms generate the optimal return-to-home path (i.e., a path that avoids obstacles such as wind turbine towers and blades, with a path length ≤ 1000m). Further considering the remaining battery power of the drone (ensuring sufficient energy for return, e.g., remaining battery percentage ≥ 30% (e.g., approximately 100Wh)), the return-to-home operation is performed. For example, the drone can first descend vertically to an altitude of 50m, then fly horizontally to a point 10m above the takeoff and landing point, and finally land vertically (landing error ≤ 0.5m). After landing, the motor control module and sensor data interface are shut down.

[0098] In step S313, the simulation main program is closed.

[0099] For example, the main simulation process sequentially shuts down the meteorological simulation, drone model, and wind turbine model modules, releasing ROS nodes and Gazebo simulation resources; finally, it outputs simulation information, which may include key indicators such as inspection coverage (≥98%), drone energy consumption (approximately 150Wh), defect identification accuracy (≥92%), and wind turbine simulation error (≤2%). At this point, the main simulation process is officially shut down, and the simulation process ends.

[0100] Next, a simulation system for aircraft-based wind turbine generator set inspection according to embodiments of the present disclosure will be described with reference to Figure 2 Figure 2 is a block diagram illustrating a simulation system for aircraft-based wind turbine generator set inspection according to embodiments of the present disclosure.

[0101] With reference to Figure 2 , the simulation system 200 for aircraft-based wind turbine generator set inspection according to embodiments of the present disclosure can include a meteorological simulation module 210, a wind turbine dynamic simulation module 220, an aircraft control simulation module 230, a wind turbine defect detection module 240, and a controller 250.

[0102] For example, the meteorological simulation module 210 can include, but is not limited to, an environment plugin subsystem (e.g., a Gazebo environment plugin subsystem), an external meteorological data interface, a meteorological data subscription-publishing submodule.

[0103] For example, the wind turbine dynamic simulation module 220 can include, but is not limited to, a basic wind turbine model (including at least blades, a tower, a nacelle, and a foundation), a wind speed and direction-blade rotation speed-rotational torque control submodule, a wind speed and direction-turbine yaw-yaw torque control submodule.

[0104] For example, the aircraft control simulation module 230 can include, but is not limited to, body and internal sensors and body external sensors. For example, the body and internal sensors can include, but are not limited to, a multi-rotor unmanned aerial vehicle body, a multi-motor control sensor, a basic positioning sensor (including a GPS, an RTK (Real Time Kinematic based on Global Satellite Navigation System), an IMU, etc.). For example, the body external sensors can include, but are not limited to, a laser radar sensor (with data publishing / subscription function), a camera sensor (with data publishing / subscription function, real-time graphical data stream display function).

[0105] Here, the sensors carried and accessed on the aircraft can also include infrared camera sensors, millimeter wave radar sensors, acoustic sensors, electromagnetic detection devices, and other types of sensors or combinations thereof that can extend the perception of the state of the wind turbine and the aircraft. By adopting a modular design architecture for the aircraft, it is possible to support the aircraft to carry multiple types of sensors, and to support multi-sensor time synchronization and data fusion, effectively avoiding the problem of data misplacement in the field application of multi-sensor collaborative solutions. In addition, a sensor expansion module is also reserved in the system architecture to realize the protocol / communication compatible access of the above-mentioned and future various types of sensors or similar devices.

[0106] ​As an example, an infrared camera can be used to detect temperature abnormalities of the blades. Specifically, since the temperature of the blades tends to change when a fault (e.g., local wear, stress concentration, etc.) occurs during the operation of the fan, the infrared camera can quickly detect the temperature abnormal area by capturing the thermal radiation distribution of the blade surface, providing an important basis for fan fault diagnosis. In addition, compared with traditional radar, millimeter wave radar has higher resolution and anti-interference ability. In the fan inspection, the millimeter wave radar can monitor the distance and relative speed between the aircraft and the fan blades in real time, providing reliable data support for obstacle avoidance and flight control of the aircraft. In addition, an acoustic sensor can be used to detect abnormal sounds during the operation of the fan, and by analyzing the sound frequency and intensity, it can be determined whether there is a fault in the internal components of the fan.

[0107] For example, the fan defect detection module 240 can include, but is not limited to, a blade defect database submodule, a model training submodule, and a model inference submodule. For example, a plurality of complex defect recognition models, such as a crack recognition model, a corrosion recognition model, and a lightning damage recognition model, can be trained based on an actual blade defect data set using, for example, a convolutional neural network algorithm, a YOLO series algorithm, etc. As an example, low-level features such as edges and textures in blade images can be extracted by using a convolutional neural network, and as the number of network layers increases, higher-level and more abstract features are gradually extracted, thereby constructing a blade defect recognition model. As another example, the YOLOv3 framework can be used to improve defect detection accuracy through multi-scale feature fusion and other techniques, which can better detect blade cracks of different sizes; the YOLOv4 framework can be used to further optimize the network structure so that the detection performance of blade cracks and other defects in a complex background is significantly improved; the YOLOv5 framework can be used to improve the lightweight and real-time performance of the model, which is suitable for unmanned aerial vehicle inspection scenarios with high real-time requirements. For example, real-time detection of fan blade cracks can be achieved based on the AI defect recognition algorithm of the YOLO framework.

[0108] Further, by applying the trained defect recognition model to the simulation system, the recognition result is fed back to the aircraft control system in real time to complete the simulation verification of the intelligent inspection whole process. Therefore, a complete and efficient intelligent inspection closed loop from data acquisition, defect recognition to control decision can be constructed.

[0109] As an example, the above-mentioned various modules can be independently encapsulated. That is, each module has a clear functional boundary and a standardized interface, flexible configuration and interaction between modules are achieved through a parameter file, thereby achieving a highly expandable architecture design. Further, in terms of the sensors used, plug-and-play of various types of sensors (e.g., visible light cameras, infrared cameras, laser radars, millimeter wave radars, etc.) is supported, and seamless integration of the sensors and the system is achieved through a standardized interface protocol. In terms of algorithm expansion, the reserved interface can conveniently access a defect identification module (e.g., an artificial intelligence (AI)-based defect identification module), thereby meeting the increasingly diversified and intelligent inspection needs of wind farms.

[0110] According to an embodiment of the present disclosure, the controller 250 can perform: by utilizing the weather simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module, and the wind turbine defect detection module, a simulation method for aircraft-based wind turbine generator inspection as described above.

[0111] Through the simulation system according to the present disclosure, a simulation environment of “large-scale unit dynamic characteristics, complex environment interaction, and actual hardware control” can be accurately achieved, a technical gap of “wind turbine non-stop inspection simulation” is filled, and safe and efficient operation and maintenance of, for example, super-large wind farms are ensured.

[0112] In addition, in the scenario of utilizing multiple unmanned aerial vehicle clusters for inspection, the method according to the present disclosure can also achieve multi-system cooperative simulation. For example, based on an extended ROS distributed architecture, through the distributed communication mechanism of ROS, the unmanned aerial vehicles can share information such as position, attitude, and task state in real time, realize cooperative operation, and thus realize cooperative inspection simulation of multiple unmanned aerial vehicles.

[0113] Specifically, in the cluster inspection scenario, multiple unmanned aerial vehicles can perform task allocation and cooperative flight according to the structural characteristics of the wind turbine and the inspection task requirements. By simulating the multi-vehicle cooperative inspection process, the cooperation strategy, communication coordination, and task allocation optimization between the unmanned aerial vehicles can be optimized, the overall inspection efficiency and reliability are improved, and reference strategies for multi-vehicle cooperative operation in actual wind power inspection are provided. Through the above integrated simulation, the actual scenario is more comprehensively simulated, and more realistic and reliable basis is provided for optimization of multi-vehicle cooperative inspection technology.

[0114] It should be noted that the operations performed by the above-mentioned various structural blocks can be similar to those described with reference to the above-mentioned various structural blocks, and thus will not be described here again. Figure 1

[0115] Hereinafter, the simulation effect of the simulation method and system for aircraft-based wind turbine generator inspection according to the present disclosure will be described with reference to Figures 4 to 12 Figures 4 to 12 ​​is a simulation process schematic diagram and simulation effect diagram of a simulation method of an aircraft-based wind turbine generator set inspection according to an embodiment of the present disclosure. Figures 4 to 7B is a simulation system interface, a UAV horizontal movement wind turbine yaw adaptive flight path point schematic diagram after simulation is completed, and a UAV vertical lifting wind turbine yaw adaptive flight path point schematic diagram. Figures 8 to 12 shows a blade defect recognition result.

[0116] Referring to Figures 4 to 7B , Figure 4 , diagram (a) and diagram (b) respectively show wind turbine rotation and yaw simulation schematic diagrams, Figure 5 shows a simulation top view of a UAV and a sensor, showing wind turbine yaw, UAV centering, two single-line laser radars, and UAV video stream. Figure 6 shows a standard operation waypoint schematic diagram planned with a wind turbine hub center as an initial point. Figure 7A shows a blade front / back surface inspection and yaw adaptive flight trajectory schematic diagram, and Figure 7B shows a blade leading / trailing edge inspection and yaw adaptive flight trajectory schematic diagram. Here, Figure 7A and Figure 7B , the coordinate axes North, East, and Height (relative to a reference ground) can correspond to the NED coordinate axes N (North), E (East), and D (ground), respectively.

[0117] Specifically, as shown in Figure 6 , in the standard operation waypoint schematic diagram planned with the wind turbine hub center as the initial point, the digital points represent point order, which indicates that the trajectory of the UAV flying in the horizontal and vertical directions before considering the yaw adaptive of the generator set is a continuous straight line. In Figure 6 , Longitude represents the longitude coordinate axis, Latitude represents the latitude coordinate axis, and Altitude represents the altitude / height coordinate axis.

[0118] In contrast, Figure 7A and Figure 7B , the digital points also represent point order, which respectively indicates that the trajectory of the UAV flying in the horizontal and vertical directions after considering the yaw adaptive of the generator set is a continuous fold. Specifically, as shown in Figure 7A , the horizontal flight waypoint trajectory of the UAV performing inspection is shown, when encountering generator set yaw, the corresponding waypoint is adaptively refreshed, changing from a straight line in the horizontal direction, for example Figure 6 , to a fold line. Similarly, as shown in Figure 7B , the vertical flight waypoint trajectory of the UAV performing inspection is shown, which can identify the yaw of the generator set by vertical lifting, the corresponding waypoint is adaptively refreshed, changing from a straight line in the vertical direction, for example Figure 6 , to a fold line.

[0119] Therefore, by Figures 6 to 7B It can be concluded that the trajectory of the UAV after the adaptive adjustment of the crew yaw in the horizontal and vertical directions changes from a continuous straight line to a continuous broken line.

[0120] Referring to Figure 8 , the effect diagram after the UAV takes a picture during the simulation process and performs defect recognition according to the picture is shown, Figure 8 The yellow part #1 in the figure indicates the defect mark identified by the wind turbine defect recognition module. For example, the mark corresponding to the identified blade crack.

[0121] Referring to Figures 9 to 12 The effect diagrams of inspecting the wind turbine according to the optimal inspection scheme obtained by applying the simulation method for wind turbine inspection based on the aircraft according to the present disclosure in the actual application scenario of the real world are shown, and Figures 9 to 12 The inspection effect diagrams of the trailing edge, front face, leading edge, and back of the blade are shown. Figures 9 to 12 The defects on the blade in the figure are represented by black to gray gradient point marks, from Figures 9 to 12 As can be seen from the figure, the method of the present disclosure can obtain effective photo data for the risk parts of the wind turbine with defects, thereby improving the inspection accuracy and efficiency.

[0122] Figure 13 is a block diagram showing a computer device 1300 according to an embodiment of the present disclosure.

[0123] Referring to Figure 13 , the computer device 1300 according to an embodiment of the present disclosure can include a processor 1310 and a memory 1320. The processor 1310 can include, but is not limited to, a central processing unit (CPU), a digital signal processor (DSP), a microcomputer, a field programmable gate array (FPGA), a system on chip (SoC), a microprocessor, an application specific integrated circuit (ASIC), etc. The memory 1320 can store computer executable instructions to be executed by the processor 1310. The memory 1320 includes a high-speed random access memory and / or a non-volatile computer readable storage medium. When the processor 1310 executes the computer executable instructions stored in the memory 1320, the simulation method for wind turbine inspection based on the aircraft as described above can be implemented.

[0124] The simulation method for aircraft-based wind turbine generator set inspection according to the embodiments of the present disclosure can be written as computer programs / instructions to form a computer program product and stored on a computer-readable storage medium. When the computer programs / instructions are executed by a processor, the simulation method for aircraft-based wind turbine generator set inspection as described above can be implemented. When the instructions in the computer-readable storage medium are executed by the processor of an electronic device / server, the electronic device / server is enabled to perform the simulation method for aircraft-based wind turbine generator set inspection as described above. Examples of the computer-readable storage medium include read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disc memory, hard disk drive (HDD), solid state drive (SSD), card memory (such as a multimedia card, a secure digital (SD) card or an extreme digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, and any other device configured to store a computer program and any associated data, data files and data structures in a non-transitory manner and provide the computer program and any associated data, data files and data structures to a processor or computer so that the processor or computer can execute the computer program. In one example, the computer program and any associated data, data files and data structures are distributed on a networked computer system, so that the computer program and any associated data, data files and data structures are stored, accessed and executed by one or more processors or computers in a distributed manner.

[0125] The simulation method and system for aircraft-based wind turbine generator set inspection according to the embodiments of the present disclosure greatly reduce the verification time of a single inspection scheme (for example, from an average of 2 days to only 0.5 days) by applying a virtual simulation environment for aircraft-based wind turbine generator set inspection, thereby greatly improving the verification efficiency of the aircraft-based inspection scheme, enabling the user to quickly compare and screen multiple inspection schemes and accelerating the optimization process of the inspection scheme.

[0126] On the other hand, the simulation method and system for aircraft-based wind turbine inspection according to the embodiments of the present disclosure can output key quantitative indicators such as system coverage and energy consumption, realize quantitative decision of the inspection scheme, provide accurate data support for optimization of the inspection scheme, and help users to analyze and make targeted adjustments to the inspection scheme based on the quantitative information, thereby improving the scientificity and rationality of the inspection scheme. In addition, by constructing an environment-turbine-aircraft simulation platform combined with defect identification, an efficient verification simulation environment for defect identification can be provided, and the defect identification effect of different models can be quickly verified through the simulation platform, thereby promoting continuous improvement and perfection of the inspection scheme, and further improving the defect identification accuracy, for example, improving the blade defect identification accuracy.

[0127] On the other hand, the simulation method and system for aircraft-based wind turbine inspection according to the embodiments of the present disclosure can simulate various risk scenarios (for example, the flexible deformation of an ultra-long blade under strong wind (for example, the maximum deformation of a blade tip reaches 1.5 m)) such as strong wind and sudden speed change of a blade in the simulation system, and comprehensively optimize the obstacle avoidance and emergency response logic of the aircraft. In actual application, the unmanned aerial vehicle inspection system optimized in this way can greatly reduce the on-site inspection accident rate (for example, by more than 60%). For example, in the simulation of a strong wind scenario, by adjusting the unmanned aerial vehicle flight attitude control algorithm, the unmanned aerial vehicle can fly more stably in a strong wind environment, effectively avoiding accidents such as collision, and ensuring the safety of inspection. Therefore, by using the method and system of the present disclosure, the safety and reliability of the inspection can be improved.

[0128] On the other hand, the simulation method and system for aircraft-based wind turbine inspection according to the embodiments of the present disclosure can reduce the development cost of the inspection scheme by using the software-in-the-loop simulation system constructed without additional hardware devices. In addition, the optimized inspection scheme can effectively reduce the number of turbine shutdowns, and each turbine can reduce the scheduled number of shutdowns (for example, 3 to 5 times) per year, thereby significantly increasing the power generation of the turbine. At the same time, the reduction in the number of shutdowns can also reduce the system operation and maintenance cost, thereby achieving cost reduction and efficiency improvement.

[0129] On the other hand, according to the software-in-the-loop simulation verification scheme of the present disclosure, a multi-dimensional evaluation index system (covering multiple key indicators such as coverage, energy consumption, response time, and positioning accuracy) is defined to systematically evaluate different aircraft inspection schemes. For example, by quantitatively comparing and analyzing the detailed data of each inspection scheme under each indicator during the simulation process, a scientific and objective basis is provided for optimization of the actual inspection scheme, thereby ensuring that the final actual application inspection scheme achieves an optimal balance in terms of efficiency, safety, and cost.

[0130] While certain embodiments of the disclosure have been disclosed and described, it will be apparent to those skilled in the art that changes and modifications can be made without departing from the conception and spirit of the disclosure, which is defined by the claims and their equivalents.

Claims

1. A method of simulating an inspection of an aircraft-based wind turbine system, the method comprising: The simulation method comprises: acquiring meteorological data, wind turbine design parameters and operation data, and aircraft inspection task data; inputting the meteorological data into a meteorological simulation module to obtain a first meteorological simulation wind field model; inputting the first meteorological simulation wind field model, the wind turbine design parameters and operation data into a wind turbine dynamic simulation module to obtain a first wind turbine dynamic simulation model; inputting the first meteorological simulation wind field model, the first wind turbine dynamic simulation model and the aircraft inspection task data into an aircraft control simulation module to control the aircraft to perform an inspection task for the wind turbine generator set, and obtaining a plurality of aircraft inspection strategy sets; inputting the plurality of aircraft inspection strategy sets into a wind turbine defect detection module to verify the effectiveness of the plurality of aircraft inspection strategy sets for wind turbine defect detection, and outputting a simulation result.

2. The simulation method of claim 1, wherein, Before the step of acquiring meteorological data, wind turbine design parameters and operation data, and aircraft inspection task data, the simulation method further comprises: in the case that the wind turbine generator set is in a running state, calling and initializing a meteorological simulation module, a wind turbine dynamic simulation module, an aircraft control simulation module and a wind turbine defect detection module.

3. The simulation method of claim 2, wherein, The step of calling and initializing the meteorological simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module and the wind turbine defect detection module comprises: calling a first running data set corresponding to the meteorological simulation module, a second running data set corresponding to the wind turbine dynamic simulation module, a third running data set corresponding to the aircraft control simulation module and a fourth running data set corresponding to the wind turbine defect detection module, and combining the first running data set to the fourth running data set to obtain a fifth running data set, based on the fifth running data set, configuring the data interfaces corresponding to the meteorological simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module and the wind turbine defect detection module for communication, so that the modules can realize data interaction through the corresponding data interfaces.

4. The simulation method of claim 1, wherein, The first wind turbine dynamic simulation model comprises a first model based on aerodynamics and a second model based on blade momentum theory, wherein the first model comprises a dynamic response model based on wind speed and direction, wind turbine yaw torque, wind turbine moment of inertia and wind turbine yaw angle, and the second model comprises a dynamic response model based on wind speed and direction, wind turbine blade aerodynamic torque and wind turbine blade speed.

5. The simulation method of claim 1, wherein, The meteorological simulation module and the wind turbine dynamic simulation module are implemented as a predetermined three-dimensional dynamic simulator, and the aircraft control simulation module is implemented as a predetermined robot operating system terminal and a predetermined flight control terminal, wherein the predetermined robot operating system terminal and the predetermined flight control terminal cooperatively perform flight steady state control for safe monitoring of unit yaw dynamic identification based on a predetermined centering algorithm and a predetermined yaw monitoring algorithm.

6. The emulation method of claim 1, wherein, The step of inputting the meteorological data into the meteorological simulation module to obtain the first meteorological simulation wind field model comprises: inputting real-time meteorological data into the meteorological simulation module to obtain a second meteorological simulation wind field model, the second meteorological simulation wind field model being a simulation wind field corresponding to the real-time meteorological data updated according to a time stamp; determining whether the second meteorological simulation wind field model can be effectively driven based on whether real-time wind speed and wind speed change rate obtained according to the second meteorological simulation wind field model meet preset conditions respectively, and taking the model determined to be able to be effectively driven as the first meteorological simulation wind field model.

7. The emulation method of claim 1, wherein, The step of inputting the first meteorological simulation wind field model and the wind turbine design parameters and operation data into the wind turbine dynamic simulation module to obtain a first wind turbine dynamic simulation model comprises: inputting the first meteorological simulation wind field model and the wind turbine design parameters and real-time operation data into the wind turbine dynamic simulation module to obtain a second wind turbine dynamic simulation model; determining consistency between first wind turbine rotating state quantity and first wind turbine yawing state quantity obtained via a topic subscription manner and second wind turbine rotating state quantity and second wind turbine yawing state quantity obtained according to the second wind turbine dynamic simulation model by using a sliding window mean algorithm; in response to the first wind turbine rotating state quantity being consistent with the second wind turbine rotating state quantity and the first wind turbine yawing state quantity being consistent with the second wind turbine yawing state quantity, taking the second wind turbine dynamic simulation model as a first wind turbine dynamic simulation model.

8. The emulation method of claim 1, wherein, Before controlling the aircraft to perform the inspection task for the wind turbine generator unit, the simulation method further comprises: determining whether the aircraft can remain at a target position under specific climate conditions and specific wind turbine operation states based on real-time acquired aircraft power state, the first meteorological simulation wind field model and the first wind turbine dynamic simulation model; in response to the aircraft being able to remain at the target position under the specific climate conditions and the specific wind turbine operation states, calibrating each component and each control parameter of the aircraft so as to control the aircraft to perform the inspection task after calibration.

9. An emulation system for wind turbine generator inspection based on an aircraft, characterized in that, The simulation system comprises: a meteorological simulation module; a wind turbine dynamic simulation module; an aircraft control simulation module; a wind turbine defect detection module; and a controller configured to implement the simulation method for aircraft-based wind turbine generator unit inspection according to any one of claims 1 to 8 by using the meteorological simulation module, the wind turbine dynamic simulation module, the aircraft control simulation module and the wind turbine defect detection module.

10. A computer device, comprising: comprise: at least one processor; at least one memory storing computer-executable instructions, wherein the computer-executable instructions, when executed by the at least one processor, cause the at least one processor to perform the simulation method for aircraft-based wind turbine generator unit inspection according to any one of claims 1 to 8.

Images