Inspection path planning method and device for operating wind turbine

Abstract

The present disclosure provides a kind of inspection path planning method and device for operating wind turbine, it is related to control technical field.The method comprises: in the online inspection stage, according to the mapping relationship generated in advance in offline stage, the maximum displacement value of blade tip under current operating condition is generated;Based on operating signal, the rigid kinematic model of blade is constructed, and the basic no-fly zone is determined;Dynamic safety radius is determined;With dynamic safety radius as inflation distance, the basic no-fly zone is outwardly inflated and handled, and the remaining space after removing dynamic no-fly zone in the preset flight space is determined as three-dimensional safe flight space;The three-dimensional inspection path of unmanned aerial vehicle is planned under the constraint of three-dimensional safe flight space, and the three-dimensional safe flight space is updated in the process of unmanned aerial vehicle flight inspection, and flight trajectory is adjusted.Based on the technical scheme, the automatic inspection path planning of operating wind turbine can be carried out in the inspection process.

Description

Technical Field

[0001] This disclosure relates to the field of control technology, and in particular to a method and apparatus for inspection path planning for operating wind turbines. Background Technology

[0002] As the global energy structure shifts towards green and low-carbon development, wind power, as a mature and cost-effective renewable energy source, is experiencing a continuous increase in installed capacity and penetration rate. Wind turbine blades are the most stress-affected and vulnerable components of the unit, requiring regular inspections to promptly identify and address any potential problems.

[0003] Currently, wind farms primarily rely on traditional manual inspections during turbine shutdown. This method requires the turbine to be completely stopped, with inspectors using high-powered telescopes for ground observation or aerial inspections via suspended platforms and ropes. To address the drawbacks of traditional manual inspections, current research and applications are focusing on automated, non-shutdown inspection methods based on mobile platforms. However, most current automated non-shutdown inspection methods involve using drones to capture panoramic views above the turbine, planning inspection trajectories for the blades, or utilizing drones equipped with high-speed cameras to capture high-speed images of rotating blades from a safe distance.

[0004] The methods described above are mostly based on simplified models when calculating blade dynamic offset or defining safety distance, which leads to a large deviation between the calculated maximum displacement and the actual situation. Furthermore, under adverse conditions such as rain, fog, strong light, or high-speed blade rotation causing image blurring, the reliability of perception decreases, making it difficult to achieve an optimal balance between safety and accuracy in the path.

[0005] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To overcome the problems existing in the related technologies, the present disclosure provides a method and apparatus for planning inspection paths for operating wind turbines, which can automatically plan inspection paths for operating wind turbines during the inspection process.

[0007] According to a first aspect of the present disclosure, a method for inspection path planning of operating wind turbines is provided. The method includes: acquiring operating signals of the wind turbine in real time during an online inspection phase, the operating signals including at least one of the following: yaw angle, rotor azimuth angle, wind speed, turbulence intensity, rotational speed, and pitch angle; querying a pre-generated mapping relationship in the offline phase based on at least one of the wind speed, turbulence intensity, rotational speed, and pitch angle acquired from the operating signals, and generating the maximum tip displacement value that the flexible deformation of the blade can achieve under the current operating conditions; constructing a rigid kinematic model of the blade based on the yaw angle and rotor azimuth angle acquired from the operating signals, and determining the blade's rigid kinematics. A basic no-fly zone is formed by the sweeping of a rigid body; the dynamic safety radius is determined based on the maximum tip displacement and the comprehensive error, which includes at least one of model error, measurement error, and engineering error; the basic no-fly zone is expanded outward using the dynamic safety radius as the expansion distance, and the remaining space after removing the dynamic no-fly zone from the preset flight space is determined as the three-dimensional safe flight space; the three-dimensional inspection path of the UAV is planned under the constraints of the three-dimensional safe flight space, and the three-dimensional safe flight space is adaptively updated based on the real-time acquired operation signals during the UAV's flight inspection, and the flight trajectory is adjusted based on the updated three-dimensional safe flight space.

[0008] According to a second aspect of the present disclosure, an inspection path planning device for operating wind turbines is provided. The inspection path planning device includes: a data acquisition module, an offline query module, a basic no-fly zone generation module, a dynamic safety radius generation module, a three-dimensional safe flight space generation module, and a path planning module. The data acquisition module is used to acquire the operating signals of the wind turbine in real time during the online inspection phase. The operating signals include at least one of the following: yaw angle, rotor azimuth angle, wind speed, turbulence intensity, rotational speed, and pitch angle. The offline query module is used to query a pre-generated mapping relationship based on at least one of the wind speed, turbulence intensity, rotational speed, and pitch angle acquired from the operating signals, and generate the maximum tip displacement value that the blade flexible deformation can achieve under the current operating conditions. The basic no-fly zone generation module is used to construct the blade based on the yaw angle and rotor azimuth angle acquired from the operating signals. The system employs a rigid body kinematics model to determine the basic no-fly zone formed by the rigid body sweep of the blades; a dynamic safety radius generation module to determine the dynamic safety radius based on the maximum tip displacement and a comprehensive error, where the comprehensive error includes at least one of model error, measurement error, and engineering error; a three-dimensional safe flight space generation module to expand the basic no-fly zone outward using the dynamic safety radius as the expansion distance to obtain the dynamic no-fly zone, and to determine the remaining space in the preset flight space after removing the dynamic no-fly zone as the three-dimensional safe flight space; a path planning module to plan the three-dimensional inspection path of the UAV under the constraints of the three-dimensional safe flight space; the dynamic safety radius generation module is also used to adaptively update the three-dimensional safe flight space based on real-time acquired operating signals during the UAV's flight inspection process; and the path planning module is also used to adjust the flight trajectory based on the updated three-dimensional safe flight space.

[0009] According to a third aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the inspection path planning method for operating wind turbines as described in the first aspect.

[0010] According to a fourth aspect of the present disclosure, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein when the computer-readable instructions are executed by the processor, they implement the inspection path planning method for operating wind turbines as described in the first aspect.

[0011] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: In this embodiment, firstly, during the online inspection phase, the operating signals of the wind turbine are acquired in real time. Based on at least one of the wind speed, turbulence intensity, rotational speed, and blade pitch angle acquired from the operating signals, a mapping relationship pre-generated in the offline phase is queried to generate the maximum blade tip displacement value that the flexible deformation of the blade can achieve under the current operating conditions. Furthermore, based on the yaw angle and rotor azimuth angle acquired from the operating signals, a rigid body kinematic model of the blade is constructed to determine the basic no-fly zone formed by the rigid body sweep of the blade. Then, based on the maximum blade tip displacement value and the comprehensive error, the dynamic safety radius is determined, and the basic no-fly zone is expanded outward using the dynamic safety radius as the expansion distance. The remaining space in the preset flight space after removing the dynamic no-fly zone is determined as the three-dimensional safe flight space. Subsequently, under the constraints of the three-dimensional safe flight space, the three-dimensional inspection path of the UAV is planned, and the three-dimensional safe flight space is adaptively updated based on the real-time acquired operating signals during the UAV's flight inspection process. The flight trajectory is also adjusted based on the updated three-dimensional safe flight space. In other words, an offline mapping relationship is first established between the maximum displacement value of the blade tip and the specific operating conditions. During the inspection process, the flexible deformation boundary of the blade is combined with the online acquired operating signals to dynamically generate a three-dimensional safe flight space. Under the constraints of this dynamic three-dimensional safe space, the path of the UAV to inspect the wind turbine is automatically planned. The deterministic motion and flexible deformation of the blade can be accurately described without relying on real-time visual perception. While ensuring the accuracy of the safety boundary, the requirements for online computing power in the online inspection process can be significantly reduced. This allows UAVs to perform efficient and safe automatic inspections of wind turbines in wind farms without shutting down the system. The wind turbines inspected by UAVs are more robust.

[0012] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0013] The accompanying drawings, which are incorporated in and form part of this disclosure, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0014] Figure 1 This is a schematic diagram of an inspection path planning system architecture for operating wind turbines, provided as an embodiment of the present disclosure.

[0015] Figure 2 This is a flowchart illustrating an inspection path planning method for operating wind turbines, provided as an embodiment of the present disclosure.

[0016] Figure 3 This is a schematic diagram of a rigid body kinematic model of a blade provided in an embodiment of this disclosure.

[0017] Figure 4This is a schematic diagram of a dynamic safety radius provided in an embodiment of this disclosure.

[0018] Figure 5 This is a schematic diagram of a drone flight path replanning provided in an embodiment of this disclosure.

[0019] Figure 6 This is a schematic diagram of the logical architecture of an inspection path planning method for operating wind turbines provided in an embodiment of this disclosure.

[0020] Figure 7 This is a hardware structure diagram of a computer device used for the inspection path planning method for operating wind turbines provided in an embodiment of this disclosure.

[0021] Figure 8 This is a schematic diagram of a path planning device for inspection of operating wind turbines, provided in an embodiment of this disclosure. Detailed Implementation

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0023] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0024] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0025] The embodiments of this disclosure will now be described in detail.

[0026] Figure 1 A schematic diagram of an inspection path planning system architecture for operating wind turbines, to which embodiments of this disclosure can be applied, is shown. Figure 1 As shown, the wind farm includes multiple operating wind turbines 101, a drone 102, and an inspection path planning device 103. The inspection path planning device 103 stores an offline mapping database 104. The drone 102 can fly and inspect the wind farm. During the inspection, the inspection path planning device 103 queries the offline mapping database 104 to find the mapping relationship between the pre-generated operating signals and the maximum tip displacement value achievable by the flexible deformation of the blades. It obtains the maximum tip displacement value achievable by the flexible deformation of the blades under the current operating conditions and constructs a rigid kinematic model of the wind turbine blades based on the operating signals to determine the basic no-fly zone formed by the rigid sweep of the blades. The inspection path planning device 103 determines the dynamic safety radius based on the maximum tip displacement value and the overall error. Using the dynamic safety radius as the expansion distance, it expands the basic no-fly zone outwards, defining the remaining space in the preset flight space after removing the dynamic no-fly zone as the three-dimensional safe flight space. Under the constraints of the three-dimensional safe flight space, the inspection path planning device 103 plans the three-dimensional inspection path of the UAV 102, and adaptively updates the three-dimensional safe flight space based on the real-time acquired operation signals during the flight inspection of the UAV 102, and adjusts the flight trajectory based on the updated three-dimensional safe flight space.

[0027] Figure 2 A flowchart illustrating an inspection path planning method for operating wind turbines provided in this disclosure is shown below. Figure 2 As shown, the method includes the following steps S201 to S206.

[0028] S201. During the online inspection phase, the operating signals of the wind turbine are acquired in real time.

[0029] The online inspection phase refers to the period during which the drone has arrived at the wind farm and begun performing blade inspection tasks.

[0030] The operating signals of the wind turbine include at least one of the following: yaw angle, rotor azimuth angle, wind speed, turbulence intensity, rotational speed, and pitch angle.

[0031] Specifically, operating signals can be read in real time through the data acquisition and monitoring control system of the wind turbine generator set.

[0032] It should be noted that the spatial position of the blades can be described by the yaw angle and the rotor azimuth angle; the flexible deformation state of the blades can be determined by wind speed, turbulence intensity, rotational speed and pitch angle.

[0033] S202. Based on at least one of the wind speed, turbulence intensity, rotational speed and blade pitch angle obtained from the operating signal, query the mapping relationship pre-generated in the offline stage to generate the maximum blade tip displacement value that the blade flexible deformation can achieve under the current operating conditions.

[0034] The offline phase refers to the stage where high-computational-cost modeling and simulation work is completed in advance before the actual UAV inspection mission begins. Operating conditions refer to the combination of operating status parameters of a wind turbine generator at a certain moment or within a certain time period (corresponding to the operating signals mentioned above). Operating status parameters are quantitative indicators describing the current operating conditions of the wind turbine, used to determine the aerodynamic and rotational loads borne by the blades, thereby affecting the flexible deformation of the blades.

[0035] Understandably, in this scheme, an offline mapping database is first generated in the offline stage. In the online line-following stage, it is not necessary to perform dynamic simulation to obtain the flexible deformation boundary. Instead, the operating signal is used as the query index to retrieve the offline mapping database and obtain the maximum tip displacement value (estimated value) under the current operating conditions.

[0036] S203. Based on the yaw angle and rotor azimuth angle obtained from the operation signal, construct a rigid body kinematic model of the blade and determine the basic no-fly zone formed by the rigid body sweep of the blade.

[0037] Among them, the rigid body kinematics model is a mathematical model used to describe the changes in position and attitude of wind turbine blades (as well as towers, nacelles, and hubs) as ideal rigid bodies in space over time. The blade rigid body is an idealized physical model assumption of wind turbine blades, which approximates the blades, which undergo elastic deformation in actual operation, as rigid objects whose shape and size remain unchanged.

[0038] In this embodiment of the disclosure, the tower, nacelle, hub and blades are first regarded as rigid geometric bodies at the path planning level. Their deterministic kinematics are described by the yaw angle and rotor azimuth angle that can be obtained in real time by the unit, so as to calculate the spatial position of the blades in operation without relying on real-time visual tracking.

[0039] Figure 3 This is a schematic diagram of a rigid body kinematic model of a blade provided in an embodiment of the present disclosure, as shown below. Figure 3 As shown, rigid geometric bodies including the wind turbine's tower, nacelle, hub, and blades are constructed in three-dimensional space.

[0040] The basic no-fly zone refers to the space occupied by the blades in space without considering flexible deformation.

[0041] It is understandable that a basic no-fly zone can provide rigid body constraints for generating a safe space for drone flight.

[0042] S204. Determine the dynamic safety radius based on the maximum displacement value of the blade tip and the comprehensive error.

[0043] The overall error includes at least one of the following: model error, measurement error, and engineering error.

[0044] The dynamic safety radius refers to the minimum distance that the basic no-fly zone (blade rigid body sweep area) needs to be extended outward at the current moment to ensure that the UAV does not collide with the rotating blade.

[0045] It is understandable that after determining the maximum displacement value of the blade tip, model errors, measurement errors, and engineering errors can be compensated to obtain a dynamic safety radius, thereby ensuring the conservatism of the safety boundary.

[0046] Figure 4 This is a schematic diagram of a dynamic safety radius provided in an embodiment of the present disclosure, as shown below. Figure 4 As shown in the figure, the left figure is the current basic no-fly zone, and the right figure is the occupied area calculated by obtaining the blade deformation in the offline stage (based on the maximum displacement value of the blade tip). The distance between the boundary of the basic no-fly zone and the boundary of the occupied area is the dynamic safety radius. It can be seen from the figure that at the same time point, the dynamic safety radius corresponding to different positions on the blade can be different.

[0047] S205. Using the dynamic safety radius as the expansion distance, the basic no-fly zone is expanded outward to obtain the dynamic no-fly zone. The remaining space after removing the dynamic no-fly zone from the preset flight space is determined as the three-dimensional safe flight space.

[0048] Among them, the preset flight space refers to the maximum three-dimensional space range in which the UAV is allowed to fly, which is a limited space area determined in the planning stage based on factors such as wind farm environment, UAV performance, and safety regulations, and serves as the global set for subsequent dynamic safety space calculation.

[0049] The remaining space refers to the spatial area obtained after removing (subtracting) the dynamic no-fly zone from the preset flight space, representing the set of all positions where the drone can safely fly at the current moment.

[0050] The three-dimensional safe flight space refers to the three-dimensional spatial region in which the UAV is permitted to fly at the current moment, meeting both safety constraints and mission requirements.

[0051] It is understandable that the above processing combines the rigid no-fly zone and the flexible deformation boundary, and the resulting three-dimensional safe flight space takes into account both the deterministic operation and the non-deterministic flexible deformation of the blades, which can provide physically consistent safety constraints for the flight of UAVs.

[0052] S206. Under the constraints of the three-dimensional safe flight space, plan the three-dimensional inspection path of the UAV, and adaptively update the three-dimensional safe flight space according to the real-time acquired operation signals during the UAV flight inspection process, and adjust the flight trajectory based on the updated three-dimensional safe flight space.

[0053] Specifically, within a three-dimensional safe flight space, a three-dimensional inspection path for the UAV can be planned with the optimization objectives of covering preset inspection target points, optimizing flight efficiency, or maintaining a safe distance. The path planning process considers task constraints such as the start point, end point, and imaging requirements to generate an executable flight trajectory.

[0054] This disclosure provides a method for planning inspection paths for operating wind turbines. First, during the online inspection phase, the operating signals of the wind turbine are acquired in real time. Based on at least one of the wind speed, turbulence intensity, rotational speed, and blade pitch angle acquired from the operating signals, a mapping relationship pre-generated in the offline phase is queried to generate the maximum tip displacement value that the flexible deformation of the blade can achieve under the current operating conditions. Furthermore, based on the yaw angle and rotor azimuth angle acquired from the operating signals, a rigid body kinematic model of the blade is constructed to determine the basic no-fly zone formed by the rigid body sweep of the blade. Then, based on the maximum tip displacement value and the comprehensive error, a dynamic safety radius is determined, and the basic no-fly zone is expanded outward using the dynamic safety radius as the expansion distance. The remaining space in the preset flight space after removing the dynamic no-fly zone is determined as the three-dimensional safe flight space. Subsequently, under the constraints of the three-dimensional safe flight space, a three-dimensional inspection path for a UAV is planned. During the UAV's flight inspection, the three-dimensional safe flight space is adaptively updated based on the real-time acquired operating signals, and the flight trajectory is adjusted based on the updated three-dimensional safe flight space. In other words, an offline mapping relationship is first established between the maximum displacement value of the blade tip and the specific operating conditions. During the inspection process, the flexible deformation boundary of the blade is combined with the online acquired operating signals to dynamically generate a three-dimensional safe flight space. Under the constraints of this dynamic three-dimensional safe space, the path of the UAV to inspect the wind turbine is automatically planned. The deterministic motion and flexible deformation of the blade can be accurately described without relying on real-time visual perception. While ensuring the accuracy of the safety boundary, the requirements for online computing power in the online inspection process can be significantly reduced. This allows UAVs to perform efficient and safe automatic inspections of wind turbines in wind farms without shutting down the system. The wind turbines inspected by UAVs are more robust.

[0055] Optionally, in the inspection path planning method for operating wind turbines provided in this embodiment of the present disclosure, before the above-mentioned S201, the offline processing steps of S207 to S209 may be included.

[0056] S207. Establish a dynamic model that includes the flexible deformation characteristics of the blades based on the structural parameters of the wind turbine.

[0057] The flexible beam model is a simplified mechanical model of a wind turbine blade. It treats the actual blade as a slender beam with continuous mass, stiffness, and damping distributions along its span (from blade root to tip). It assumes that deformation primarily occurs in flapping (perpendicular to the chord length) and yaw (parallel to the chord length), neglecting torsional deformation. This model is a core component of the flexible dynamics model and is used for offline simulation of the elastic deformation of the blade under aerodynamic and rotational loads.

[0058] Specifically, the blades are considered as flexible beams distributed along the span, and a dynamic model of the blades' flexibility can be established based on the structural parameters of the wind turbine.

[0059] The structural parameters of a wind turbine can be obtained through its design parameters and / or on-site 3D scanning data. These parameters include at least one of the following: blade shape, mass distribution, and stiffness distribution. The design parameters include parametric models of blade shape, hub radius, tower shape, nacelle center coordinates, and hub center coordinates. The on-site 3D scanning data consists of point cloud data obtained after UAV scanning of the wind turbine, followed by registration, denoising, and surface reconstruction.

[0060] S208. Enumerate multiple operating conditions, perform offline simulation of the dynamic model based on multiple operating conditions, and calculate the maximum displacement envelope that the blade can achieve after flexible deformation under each operating condition.

[0061] It should be noted that, at the path planning level, the tower, nacelle, hub and blades are first regarded as rigid geometric bodies, and the deterministic dynamics of the unit are characterized by the yaw angle and rotor azimuth angle obtained in real time by the unit.

[0062] Specifically, key operating parameters such as wind speed, turbulence intensity, rotational speed, and blade pitch angle can be sampled in combination to form various design load conditions. For each set of conditions, the corresponding equivalent load is applied to the dynamic model, the blade deformation time history is solved, and the maximum tip displacement value is extracted.

[0063] S209. Establish a mapping relationship from the online measurable operating signals corresponding to various operating conditions to the corresponding maximum displacement envelope, and store it as an offline mapping database.

[0064] It is understandable that by associating and storing the working condition vector with the calculated maximum blade tip displacement value, an offline mapping database can be constructed, thereby reducing the online calculation process to a simple query / interpolation operation.

[0065] Based on this scheme, the complex aeroelastic deformation calculation of the blade is completed offline. The offline simulation fully considers the aeroelastic coupling effect of the blade and can accurately calculate the maximum displacement envelope of the blade under various operating conditions of the wind turbine. When the UAV is inspecting online, it only needs to obtain the flexible deformation boundary of the blade by looking up the table and / or interpolation, without the need for high-cost dynamic calculations online, which significantly reduces the system complexity and computing power requirements.

[0066] Optionally, in the inspection path planning method for operating wind turbines provided in the embodiments of this disclosure, the above-mentioned S208 specifically includes S208a to S208d.

[0067] S208a. Simplify the blade into a flexible beam model distributed along the spanwise direction, and determine the mass distribution, damping distribution, and stiffness distribution of the flexible beam model.

[0068] Stiffness distribution is a physical quantity describing the ability of wind turbine blades to resist bending deformation along their spanwise cross-sections, and is one of the core input parameters of the flexible beam model. It comprehensively reflects the changes in the elastic modulus and cross-sectional geometry (moment of inertia) of the blade material along the blade length.

[0069] Specifically, in the flexible dynamics model, the blade is considered as a variable cross-section Euler-Bernoulli beam distributed along the spanwise direction, and the wind turbine blade is mapped along the spanwise coordinates. Parameterization The blade length is represented, and then the mass distribution per unit length, equivalent damping distribution, and equivalent bending stiffness distribution of the blade are obtained.

[0070] It is understandable that by simplifying the blade into a flexible beam model distributed along the span, a structural simplification basis can be provided for subsequent dynamic modeling, so that the complex blade continuum can be characterized by a finite number of parameters, reducing the degree of freedom of the model, and ensuring the engineering accuracy of the blade deformation calculation while retaining the variable cross-section characteristics of the blade along the span.

[0071] S208b: The first K natural modes and natural frequencies of the blade are solved using the modal superposition method, and the continuous deformation of the blade is converted into a reduced-order dynamic system with modal coordinates as variables based on the solution results.

[0072] Where K is a positive integer.

[0073] It should be noted that the Modal Superposition Method (MSM) is a structural dynamics reduction technique used to approximate the infinite-degree-of-freedom vibration problem of a continuum (such as the blade in this disclosure). A reduced-order dynamic system refers to a dynamic model that transforms the original blade continuous elastic body (described by partial differential equations) with infinitely many degrees of freedom into a system of ordinary differential equations with a finite number of degrees of freedom, using the MSM method. The state variables of this system are modal coordinates, where K is the truncated modal order, and its motion is controlled by a set of mutually decoupled (or weakly coupled) second-order ordinary differential equations. This system of ordinary differential equations with a finite number of modal coordinates significantly reduces computational complexity.

[0074] Specifically, the method of analyzing beam deformation from related technologies is introduced into the construction of the safe flight domain of UAVs. The modal superposition method can be used to establish the bending vibration differential equation of the blade based on beam theory.

[0075] For example, using the Euler-Bernoulli beam assumption, the bending vibration of the blade can be described by the partial differential equation shown in formula (1).

[0076] ;Formula (1) in, Represents the spanwise coordinates of any point on the blade. Indicates the blade in spanwise coordinates Mass distribution per unit length at a given location Indicates time, Indicates time blade in span coordinates The deflection field at that location, Indicates w to Find the second derivative. express right Find the first derivative. Indicates the blade in spanwise coordinates Equivalent damping distribution at the location, Indicates the blade in spanwise coordinates The equivalent bending stiffness distribution at the location. Indicates time Span coordinates acting on the blade Equivalent distributed external load at the location.

[0077] The equivalent distributed external loads include: equivalent aerodynamic loads and equivalent rotational dependent loads.

[0078] It is understandable that establishing partial differential equations for the bending vibration of blades based on beam theory can accurately express the physical relationship between aerodynamic loads, rotational loads and blade structural response in mathematical form, providing a complete dynamic control equation for subsequent blade deformation calculation. This allows the aeroelastic coupling effect of the blade to be fully considered in offline simulation, avoiding displacement calculation errors caused by simplified models.

[0079] Afterwards, the modal superposition method can be used to reduce the order. First, the first K natural modes and natural frequencies of the blade are solved; then, based on the natural modes and natural frequencies, the blade deflection is approximated as a linear combination of modal coordinates based on formula (2).

[0080] ;Formula (2) Where K represents the modal order used. Indicates the modal order index. , Indicates the first First-order natural mode shape parameters, Indicates the first The coordinates of the first intrinsic mode (generalized coordinates that change with time).

[0081] It is understandable that by using the modal superposition method to solve for the natural modes and natural frequencies, and then transforming the continuous partial differential equations into a modal coordinate time-domain dynamic system with K finite degrees of freedom, the computational complexity of offline simulation can be significantly reduced while maintaining the main dynamic characteristics of blade deformation. This makes it possible to perform rapid simulations under a large number of design load conditions and provides a computationally feasible way to establish an offline displacement mapping database covering all working conditions.

[0082] S208c: Based on the wind speed, turbulence intensity, rotational speed and pitch angle under each operating condition, determine the equivalent distributed load acting on the blade.

[0083] For example, based on the concept of DLC (Design Load Case) or an equivalent set of load cases, key operating quantities such as wind speed, turbulence intensity, rotational speed and pitch angle are sampled in combination, and the aerodynamic load and rotation-related load are equivalent to distributed excitation in each set of designed operating conditions.

[0084] For example, for each set of operating conditions, the time is calculated based on the aerodynamic model. Span coordinates on the blade Equivalent distributed load at the location , Indicates wind speed. Indicates turbulent velocity. Indicates the speed of the wind turbine. Indicates the propeller pitch angle.

[0085] It is understandable that by using real-time measurable or simulable operating parameters such as wind speed, turbulence intensity, and rotational speed, the equivalent distributed load acting on the blade can be determined, and the external aerodynamic environment and operating state can be correlated with blade deformation. This allows offline simulation to cover various actual operating scenarios from low wind speed to high wind speed and from steady state to turbulence, ensuring the integrity of the operating conditions for calculating the displacement envelope.

[0086] S208d: Apply the equivalent distributed load to the reduced-order dynamic system and solve the time history of blade deformation; determine the maximum displacement at any specified position on the blade tip or blade from the time history, and use the maximum displacement as the maximum displacement envelope under each operating condition.

[0087] In this context, the time history refers to the complete record or functional relationship of how a physical quantity (such as modal coordinates, tip displacement, blade deflection, etc.) changes over time. It typically uses time as the independent variable to describe the dynamic evolution of the physical quantity from the initial moment to the final moment. The maximum displacement envelope refers to the upper bound of the maximum possible displacement of the blade at all positions within a complete time window, under a given design load condition (determined by parameters such as wind speed, turbulence intensity, rotational speed, and blade pitch angle).

[0088] Specifically, the equivalent distributed load Applying to the reduced-order dynamic system, the time history of the modal coordinates is solved. Based on the time history of the modal coordinates, the mode shape function, and the above formula (2), the blade deflection field is reconstructed. From the reconstructed blade deflection field, the displacement value of the blade tip position at each time moment is extracted, and the maximum value is taken as the maximum displacement value of the blade tip under this working condition.

[0089] For example, the maximum displacement value of the blade tip can be calculated based on formula (3).

[0090] ;Formula (3) in, This represents the upper limit of the maximum displacement of the blade tip (a function of time and operating parameters). Indicates the length of the simulation time window. This indicates that the peak value is taken within the time window. This indicates modulo or absolute value operations. This represents the tip displacement vector (a function of time and operating parameters). , Indicates time Flapwise displacement of leaf tips Indicates time Edgewise displacement.

[0091] It should be noted that, in order to ensure the conservatism of the upper limit of displacement, in the embodiments of this disclosure, multiple turbulent random seeds can be used to repeatedly calculate the same working condition, and the maximum value or high quantile of the tip displacement peak value can be taken as the displacement envelope of the working condition.

[0092] It is understandable that by applying an equivalent distributed load to the reduced-order dynamic system and solving the time history, the deformation response of the blade within the complete time window can be obtained; extracting the maximum value of the tip displacement as the displacement envelope under each working condition provides physically consistent upper bound data for the expansion of the online safety space, avoiding the high cost of online real-time deformation calculation.

[0093] Based on this scheme, a flexible beam model and modal superposition method are adopted. While ensuring the accuracy of calculation, efficient order reduction can be achieved. It can accurately describe the distribution of mass, stiffness and damping of the blade along the spanwise direction, and fully consider the aeroelastic coupling effect of the blade. This provides a reliable data foundation for determining the safety boundary of UAV flight inspection. At the same time, offline pre-calculation avoids the high cost of online dynamic simulation, which significantly reduces the dependence of the online inspection system on real-time computing power.

[0094] Optionally, in the inspection path planning method for operating wind turbines provided in the embodiments of this disclosure, the above-mentioned S202 may specifically include the following S202a to S202c.

[0095] S202a. Using at least one of the wind speed, turbulence intensity, rotational speed, and pitch angle from the online real-time acquired operating signals as the query index, retrieve the offline mapping database.

[0096] For example, the offline mapping database uses work condition vectors Use it as an index to store the corresponding maximum displacement value of the blade tip.

[0097] It should be noted that when certain operating signals are unavailable, or the obtained values ​​are invalid, the index dimensions can be reduced accordingly to query / interpolate and obtain the maximum displacement value of the blade tip.

[0098] S202b: When the real-time operating signal matches the discrete operating point in the offline mapping database, the tip displacement value in the corresponding maximum displacement envelope is determined as the tip maximum displacement value.

[0099] S202c: When the real-time operating signal is between discrete operating points in the offline mapping database, the maximum blade tip displacement value under the current operating signal is determined by interpolation.

[0100] For example, the maximum position value of the blade tip under the current operating signal can be calculated by interpolation operation based on the following formula (4).

[0101] ;Formula (4) in, Indicates time Estimated maximum blade tip displacement Indicates an offline mapping database. This represents the database interpolation / matching operator. Indicates time wind speed, Indicates time turbulence intensity, Indicates time The wind turbine speed, Indicates time The pitch angle.

[0102] It should be noted that, in the above data interpolation, this disclosure takes into account wind speed, turbulence intensity, rotational speed, and propeller pitch angle for interpolation.

[0103] Based on this scheme, the actual working conditions of online inspection can be quickly matched with the offline mapping database, avoiding the need for complex dynamic solutions during UAV online inspection. The maximum blade tip displacement value under continuous working conditions can be obtained simply by looking up tables and interpolation, ensuring lightweight and accurate real-time response. This can improve the system's robustness and deployability under conditions of rain, fog, strong light, and high-speed rotation that cause imaging degradation.

[0104] Optionally, in the inspection path planning method for operating wind turbines provided in the embodiments of this disclosure, the above-mentioned S203 may specifically include the following S203a to S203c.

[0105] S203a. Based on the three-dimensional geometric model and operating signals of the wind turbine constructed in the offline stage, construct a coordinate transformation chain from the ground inertial coordinate system through the nacelle coordinate system rotating with the yaw angle to the wind turbine coordinate system rotating with the wind turbine azimuth angle.

[0106] Specifically, during the offline phase, a set of rigid body geometries for the wind turbine is constructed based on structural parameters. This set of rigid body geometries includes mesh or parametric surface data of the tower, nacelle, hub, and blades. Among them, the blade geometry data is used to determine the local coordinates of any point on the blade; the geometry data of static obstacles such as the tower and nacelle are used to subsequently merge with the blade swept area to form a complete basic no-fly zone.

[0107] To facilitate the description of motion and occupancy, this disclosure defines four coordinate systems: a ground inertial coordinate system {G}, a nacelle yaw coordinate system {N}, a rotor coordinate system {R}, and a blade local coordinate system {B}. The transformation relationships between these four coordinate systems are constructed, and the transformation chain from the local coordinates of a point on the blade to the ground inertial coordinate system is described by a rotation matrix. Specifically, the Z-axis of the ground inertial coordinate system {G} points vertically upwards; the nacelle yaw coordinate system {N} rotates by a yaw angle relative to the ground inertial coordinate system {G} around its Z-axis; and the rotor coordinate system {R} rotates by an azimuth angle relative to the nacelle yaw coordinate system {N} around its principal axis. The blade local coordinate system {B} is fixed to each blade and is used to describe the scanned blade geometry. The coordinate transformation chain matrix can be represented as follows: and . This represents the rotation matrix from the ground inertial coordinate system {G} to the cabin yaw coordinate system {N}. ; This represents the rotation matrix from the yaw coordinate system {N} of the engine room to the wind turbine coordinate system {R}. Indicates the yaw angle. Indicates the azimuth angle of the wind turbine. The transformation from the blade local coordinate system {B} to the rotor coordinate system {R} includes a fixed phase offset determined by the number of blades. This transformation, combined with the rotation of the rotor coordinate system {R} relative to the nacelle yaw coordinate system {N}, is unified by a rotation matrix. describe.

[0108] S203b. Based on the three-dimensional geometric model of the wind turbine, determine the local coordinates of any point on the blade in the wind turbine coordinate system. Using the coordinate transformation chain, transform the local coordinates of any point on the blade to the nacelle yaw coordinate system and then to the ground inertial coordinate system to obtain the instantaneous position coordinates of any point in the ground inertial coordinate system.

[0109] Specifically, for the first The fixed geometric position of any point on a blade. It can be uniquely determined by the combination of the chordal and thickness cross sections of the blade and its spanwise position, and is usually given by the blade's three-dimensional point cloud data or geometric model.

[0110] For example, the instantaneous position of each point on each blade of the wind turbine in the ground inertial coordinate system can be obtained by the following formula (5).

[0111] ;Formula (5) in, Indicates the blade serial number. It is a positive integer. This represents the total number of blades in a wind turbine. For a three-bladed wind turbine, , Indicates time No. spanwise coordinates on each blade The position vector in the ground inertial coordinate system at that location. Indicates time The position vector of the hub center (or the origin of the equivalent wind turbine coordinate system) in the ground inertial coordinate system. It can be a constant or include a time variable that includes tower top vibration. This represents the rotation matrix from the cabin yaw coordinate system to the ground inertial coordinate system. Indicates time Yaw angle, Rotation matrix from the wind turbine coordinate system to the nacelle yaw coordinate system Indicates time The azimuth angle of the wind turbine, This means first rotating the geometric coordinates of the blade point cloud data from the wind turbine coordinate system to the yaw coordinate system, and then rotating it to the ground inertial coordinate system. Represents the spanwise coordinates of the blade The position vector of the point at that location in the local coordinate system of the blade.

[0112] Taking a three-bladed wind turbine as an example, the phase difference between two adjacent blades, i-th and i+1-th blades, is 2π / 3. If we take a two-bladed wind turbine as an example, the phase difference between the two blades is 2π / 2.

[0113] S203c. Within a planned time window, determine the volume set formed by the rigid body sweep of the blades and merge it with static obstacles to obtain the basic no-fly zone.

[0114] Alternatively, the volume set formed by the rigid body sweep of the blade can be obtained in the following two ways.

[0115] Method 1: Using the wind turbine coordinate system as a reference, adjust the time... wind turbine azimuth exist Discrete sampling is performed within the range, the instantaneous position of all points on the blade surface is calculated at each sampling time, and the set of instantaneous positions at each sampling time is subjected to a union operation.

[0116] Method 2: Discretely sample time in the ground inertial coordinate system, calculate the instantaneous position of all points on the blade surface at each sampling time, and perform a union operation on the set of instantaneous positions at each sampling time.

[0117] For example, based on formula (6), the set of rigid body swept volumes of the blades is combined with the geometry of the tower and the nacelle to obtain the basic no-fly zone.

[0118] ;Formula (6) in, Indicates the basic no-fly zone. Indicates the cabin area. Indicates the tower area, This represents the volume set formed in three-dimensional space by the movement trajectory of the blades along the azimuth angle of the wind turbine.

[0119] It is understandable that a basic no-fly zone is the space that drones must avoid without considering flexible deformation.

[0120] Based on this scheme, when using UAVs for online inspection of operating wind turbines, there is no need to rely on real-time visual tracking of the wind turbines. The no-fly zone formed by the rigid body sweep of the wind turbine blades can be accurately calculated using only the yaw angle and rotor azimuth angle from the real-time acquired operating parameters of the wind turbines. This provides precise geometric constraints for the dynamic expansion of the no-fly zone by the UAVs, reduces the computational load of online inspection, and improves the robustness of UAVs for online inspection of operating wind turbines.

[0121] Optionally, in the inspection path planning method for operating wind turbines provided in the embodiments of this disclosure, the above-mentioned S204 may specifically include the following S204a.

[0122] S204a. The dynamic safety radius is obtained based on the maximum displacement value of the blade tip, the safety amplification factor, and the fixed safety margin.

[0123] Among them, the safety amplification factor is used to compensate for the deviation between simulation and actual measurement, and the fixed safety margin is used to cover engineering factors such as the positioning error, control error and communication delay error of the UAV.

[0124] For example, based on formula (7), the safety amplification factor and the fixed safety margin, the estimated value of the maximum displacement of the blade tip is converted into the dynamic safety radius.

[0125] ; Formula (7) in, Indicates a point in time Dynamic safety radius at time, Indicates the safety amplification factor. , This indicates a fixed safety margin, expressed in meters.

[0126] Furthermore, the time point is determined in S204a above. After determining the dynamic safety radius, in S205 above, the time point can be determined based on the following formulas (8) and (9). A three-dimensional safe flight space.

[0127] ;Formula (8) ;Formula (9) in, Indicates time Dynamic no-fly zones Indicates time Basic no-fly zone, Indicates time spherical expansion operator; Indicates time A three-dimensional safe flight space This indicates the preset flight space.

[0128] Based on this scheme, the theoretically calculated upper limit of displacement can be transformed into a practically usable safe radius. By introducing a safety amplification factor and a fixed margin, model errors, sensor measurement errors, short-term gusts, and unmodeled disturbances can be uniformly incorporated into the safety margin. This avoids excessive conservatism while ensuring the safety of UAV flight and wind turbine operation, thus achieving a balance between safety and operational efficiency.

[0129] Optionally, in the inspection path planning method for operating wind turbines provided in the embodiments of this disclosure, the above-mentioned S206 may specifically include the following S206a to S206c.

[0130] S206a. Set the start and end points of the inspection task, as well as the sequence of target points on the blade surface that need to be imaged, and set the imaging requirements.

[0131] The imaging requirements include the desired viewing distance, the desired angle of incidence, and the shooting distance range.

[0132] Specifically, each target point is accompanied by the desired shooting distance and angle of incidence range.

[0133] S206b. Construct a comprehensive optimization objective function within a three-dimensional safe flight space.

[0134] The comprehensive optimization objective function is a weighted combination of path length cost and safety distance cost.

[0135] S206c. Using a path search or sampling planning algorithm, with the goal of minimizing the comprehensive optimization objective function, the algorithm searches for the optimal path from the starting point to the ending point, passing through the observable areas near each target point in sequence, and satisfying the target constraints.

[0136] The target constraints include at least one of the following: the line of sight from waypoints on the path to the corresponding target point does not intersect the dynamic no-fly zone; the distance from waypoints to the target point is within the set shooting distance range; and the angle between the line of sight and the normal to the blade surface meets the desired angle of incidence requirement.

[0137] For example, sampling planning algorithms such as RRT* (Rapidly-exploring Random Tree Star) or PRM (Probabilistic Roadmap Method) can be used to search for the optimal path under the guidance of the comprehensive optimization objective function. It should be noted that adding line-of-sight constraints during the search process can ensure that the line of sight from waypoints on the path to the corresponding target point does not intersect with the basic no-fly zone, that is, satisfy the constraint conditions shown in formula (11).

[0138] For example, the comprehensive optimization objective function can be constructed by minimizing the constraints based on the comprehensive optimization objective function indicated by formula (10) and the following formula (11).

[0139] ; Formula (10) ; Formula (11) in, Indicates the location of the drone at time t. This represents the comprehensive optimization objective function. This represents the comprehensive optimization objective function that minimizes the UAV's position at time t. , The weighting coefficients represent the path length cost. The weighting coefficients represent the cost of maintaining a safe distance. This represents the dynamic no-fly zone at time t. This represents the safety penalty function. This represents the position of the target point in the ground inertial coordinate system at time t. Indicates the location of the drone To the target point The set of line segments of sight, This indicates that the following condition is met. This indicates that drone collision detection uses visual sensors / millimeter-wave radar / LiDAR to obtain the drone's position (x) and the wind turbine's position (y). Then, it performs line of sight (LOS) intersection calculation, meaning that the path from x to y is safe, flyable, and unobstructed.

[0140] In this embodiment of the disclosure, after obtaining the flight path, the flight path can be discretized into a time-stamped sequence of waypoints (including position, speed, heading / gimbal pointing and trigger time if necessary), which can be used as executable mission instructions for the UAV flight control system.

[0141] Based on this scheme, inspection efficiency is guaranteed at the cost of path length, safety distance is guaranteed to keep away from risk boundaries at the cost of safe distance, and line-of-sight constraints guarantee shooting quality. This achieves comprehensive optimization of safety and efficiency within a dynamic safe space, forming an executable, imageable, and verifiable inspection trajectory for drones.

[0142] Optionally, in the inspection path planning method for operating wind turbines provided in the embodiments of this disclosure, the above-mentioned S206 may specifically include S206d to S206f as described below.

[0143] S206d: During the inspection process of the UAV, the operating signals of the wind turbine are continuously read at a preset cycle.

[0144] Optionally, if the wind turbine's operating signal is not read in the next preset cycle, it can retreat to a safe area outside the operating wind turbine, hover, return, or continue flying in the three-dimensional safe flight space of the previous cycle. Pre-setting or real-time manual intervention can be performed as needed, and this disclosure does not specifically limit this.

[0145] S206e: Within each cycle, the maximum blade tip displacement value is re-determined based on the latest operating signals, and the three-dimensional safe flight space is updated.

[0146] For example, within each period, discrete time points are defined. ,in, Indicates the start time of the period. Indicates the control / navigation update cycle. This represents the discrete-time step index, at the discrete time point. At this point, real-time operation signals are obtained, and each time point is executed based on the above formulas (4), (7) to (9). The update operation yields , , , , ,in, Indicates a point in time The working condition vector, Indicates a point in time The estimated maximum displacement of the blade tip. Indicates a point in time The dynamic safety radius, Indicates a point in time Dynamic no-fly zones; Indicates a point in time A three-dimensional safe flight space.

[0147] S206f: When it is determined that the updated three-dimensional safe flight space meets the path replanning conditions, the current position of the UAV is taken as the new starting point, the subsequent path is replanned based on the updated three-dimensional safe flight space, and a new obstacle avoidance trajectory instruction is generated.

[0148] Among them, the path replanning conditions meet any of the following: the operating state crosses the working condition interval, causing the dynamic safety radius to be greater than the preset radius; the change in yaw angle or wind turbine azimuth angle causes the drift caused by blade sweeping to be greater than the preset angle; when the updated safe flight space is detected to intersect with the currently unexecuted path; and the distance between the UAV position and the boundary is less than the preset distance.

[0149] For example, the preset radius is equal to twice the blade chord length or the tower diameter, the preset angle can be set to 5°, and the preset distance can be set to twice the maximum length of the drone.

[0150] For example, the drone in the 1st month can be predicted based on the following formula (12). n +1 time point 3D position waypoints And based on location waypoints Whether it is located within the updated three-dimensional safe flight space, and whether the drone's position meets the conditions for path replanning.

[0151] ; Formula (12) in, Indicates the first Time points UAV 3D location waypoints Indicates the first Time points The speed indicated or estimated by the drone speed command. A database interpolation / matching operator that maps the current state to an estimate of the maximum tip displacement.

[0152] When performing local replanning, the current position of the UAV is used as the new starting point, and the subsequent path is replanned based on the updated safe flight space, and a new waypoint sequence is generated.

[0153] When implementing the retreat strategy, the drone is guided to a preset safe waiting point. This waiting point is located inside the updated safe flight space and far from the updated dynamic no-fly zone. Subsequent flight segments are then regenerated after the operating conditions stabilize.

[0154] Figure 5 This is a schematic diagram of a drone flight path replanning provided in an embodiment of the present disclosure, such as... Figure 5 As shown in (a), this is the path before planning. Figure 5(b) in the diagram represents the replanned path. (Combined with...) Figure 4 and Figure 5 As can be seen, the blades rotate over time, and the basic no-fly zone and occupied area change over time. After replanning, the UAV can still complete tasks such as tower inspection, hub hovering, and blade area inspection. However, its flight trajectory has been dynamically adjusted according to the current environmental conditions, operation location, and blade occupied area distribution. The inspection path is more in line with the wind turbine structure and real-time operating conditions. The slow sweeping section and detour section are also more reasonable, thereby improving the inspection efficiency of wind turbine units while avoiding blade occupied areas.

[0155] Based on this scheme, the system can respond to changes in operating conditions in real time, dynamically adjust the safety space, and replan the path locally when necessary. This forms a complete closed loop of offline high-fidelity pre-simulation, online signal matching and updating, and local replanning when necessary, enabling adaptive obstacle avoidance during flight and maintaining flight safety even in complex environments.

[0156] Figure 6 This is a schematic diagram of the logical architecture of an inspection path planning method for operating wind turbines provided in an embodiment of this disclosure, as shown below. Figure 6 As shown, the process includes an offline phase and an online inspection phase. In the offline phase, the aeroelastic simulation module establishes a blade beam model, performs offline aeroelastic simulation under DLC, extracts the maximum tip displacement under various operating conditions, and constructs an offline mapping database of maximum displacement and operating signals. In the online inspection phase, the operating signals of the wind turbine are read online. On one hand, based on the operating signals, a three-dimensional geometric model of the wind turbine is obtained, yaw and rotor rotation models are established, blade rotation sweep occupancy is calculated, and a basic no-fly zone is generated. On the other hand, based on the operating signals, the upper limit of the current maximum displacement is matched by querying / interpolating the offline mapping database, a safety margin is introduced, and the dynamic safety radius is calculated. Then, based on the dynamic safety radius, the basic no-fly zone is expanded to generate a dynamic no-fly zone. The dynamic no-fly zone is subtracted from the planned space to obtain a three-dimensional safe flight space. Path planning and pre-set shooting optimization are performed within the three-dimensional safe flight space to generate executable tracks and execute inspection tasks. Operating signals are periodically read, and the three-dimensional safe flight space is updated based on the offline mapping database to determine whether the updated three-dimensional safe flight space meets the path replanning conditions. If the conditions are not met, continue to periodically read the operation signals to update the three-dimensional safe flight space; if path planning is performed within the updated three-dimensional safe flight space.

[0157] Corresponding to the embodiments of the foregoing methods, this disclosure also provides embodiments of the apparatus and the computer equipment on which it is applied.

[0158] Embodiments of the disclosed device can be applied to computer equipment, such as servers or terminal devices. The device embodiments can be implemented through software, hardware, or a combination of both. Taking software implementation as an example, as a logically defined inspection path planning device for operating wind turbines, it is formed by the processor responsible for the inspection path planning of operating wind turbines loading the corresponding computer program instructions from non-volatile memory into memory for execution. From a hardware perspective, such as... Figure 7 The diagram shown is a hardware structure diagram of a computer device housing an inspection path planning device for operating wind turbines, according to an embodiment of this disclosure. Except for... Figure 7 In addition to the processor 710, memory 730, network interface 720, and non-volatile memory 740 shown, the server or electronic device in the embodiment where the inspection path planning method for operating wind turbines is located may also include other hardware depending on the actual function of the computer device, which will not be described in detail here.

[0159] like Figure 8 As shown, Figure 8This disclosure provides an inspection path planning device for operating wind turbines. The inspection path planning device 103 includes: a data acquisition module 1031, an offline query module 1032, a basic no-fly zone generation module 1033, a dynamic safety radius generation module 1034, a three-dimensional safe flight space generation module 1035, and a path planning module 1036. The data acquisition module 1031 is used to acquire the operating signals of the wind turbine in real time during the online inspection phase. The operating signals include at least one of the following: yaw angle, rotor azimuth angle, wind speed, turbulence intensity, rotational speed, and pitch angle. The offline query module 1032 is used to query a pre-generated mapping relationship based on at least one of the wind speed, turbulence intensity, rotational speed, and pitch angle acquired from the operating signals, and generate the maximum tip displacement value that the flexible deformation of the blade can achieve under the current operating conditions. The basic no-fly zone generation module 1033 is used to generate the maximum tip displacement value that the flexible deformation of the blade can achieve under the current operating conditions based on the yaw angle and rotor azimuth angle acquired from the operating signals. The system includes: an azimuth angle module for constructing a rigid body kinematic model of the blade and determining the basic no-fly zone formed by the rigid body sweep of the blade; a dynamic safety radius generation module 1034 for determining the dynamic safety radius based on the maximum tip displacement and comprehensive error, where the comprehensive error includes at least one of model error, measurement error, and engineering error; a three-dimensional safe flight space generation module 1035 for expanding the basic no-fly zone outward using the dynamic safety radius as the expansion distance to obtain the dynamic no-fly zone, and determining the remaining space after removing the dynamic no-fly zone from the preset flight space as the three-dimensional safe flight space; a path planning module for planning the three-dimensional inspection path of the UAV under the constraints of the three-dimensional safe flight space; the three-dimensional safe flight space generation module 1035 for adaptively updating the three-dimensional safe flight space based on real-time acquired operating signals during the UAV's flight inspection process; and a path planning module 1036 for adjusting the flight trajectory based on the updated three-dimensional safe flight space.

[0160] Optionally, the inspection path planning device further includes: a dynamic model generation module, an offline simulation module, and an offline mapping database generation module; the dynamic model generation module is used to establish a dynamic model including the flexible deformation characteristics of the blades based on the structural parameters of the wind turbine; the offline simulation module is used to enumerate multiple operating conditions, perform offline simulation on the dynamic model based on multiple operating conditions, and calculate the maximum displacement envelope that the blades can achieve after flexible deformation under each operating condition; the offline mapping database generation module is used to establish the mapping relationship from the online measurable operating signals corresponding to various operating conditions to the corresponding maximum displacement envelope, and store it as an offline mapping database.

[0161] Optionally, the offline simulation module is specifically used for: simplifying the blade into a flexible beam model distributed along the spanwise direction, and determining the mass distribution, damping distribution, and stiffness distribution of the flexible beam model; solving for the first K natural modes and natural frequencies of the blade using the modal superposition method, and converting the continuous deformation of the blade into a reduced-order dynamic system with modal coordinates as variables, where K is a positive integer; determining the equivalent distributed load acting on the blade based on the wind speed, turbulence intensity, rotational speed, and pitch angle under each operating condition; applying the equivalent distributed load to the reduced-order dynamic system and solving for the time history of blade deformation; determining the maximum displacement value at the blade tip or any specified position on the blade from the time history, and using the maximum displacement value as the maximum displacement envelope under each operating condition.

[0162] Optionally, the offline query module is specifically used to: retrieve the offline mapping database using at least one of the wind speed, turbulence intensity, rotational speed, and blade pitch angle from the online real-time acquired operating signals as the query index; when the real-time operating signal matches a discrete operating point in the offline mapping database, determine the blade tip displacement value in the corresponding maximum displacement envelope as the maximum blade tip displacement value; when the real-time operating signal is between discrete operating points in the offline mapping database, determine the maximum blade tip displacement value under the current operating signal through interpolation.

[0163] Optionally, the dynamic safety radius generation module is specifically used to: obtain the dynamic safety radius based on the maximum tip displacement value, the safety amplification factor, and the fixed safety margin; wherein, the safety amplification factor is used to compensate for the deviation between simulation and actual measurement, and the fixed safety margin is used to cover the positioning error, control error, and communication delay error of the UAV.

[0164] Optionally, the basic no-fly zone generation module is specifically used for: constructing a coordinate transformation chain from the ground inertial coordinate system through the nacelle yaw coordinate system rotating with the yaw angle to the wind turbine coordinate system rotating with the wind turbine azimuth angle, based on the three-dimensional geometric model and operating signals of the wind turbine built in the offline stage; determining the local coordinates of any point on the blade in the wind turbine coordinate system based on the three-dimensional geometric model of the wind turbine; using the coordinate transformation chain, transforming the local coordinates of any point on the blade to the nacelle yaw coordinate system and then to the ground inertial coordinate system to obtain the instantaneous position coordinates of the arbitrary point in the ground inertial coordinate system; determining the volume set formed by the rigid body sweep of the blade within a planned time window, and merging it with static obstacles to obtain the basic no-fly zone formed by the rigid body sweep of the blade.

[0165] Optionally, the path planning module is specifically used for: setting the starting point, ending point, and sequence of target points on the blade surface to be imaged for the inspection task; setting imaging requirements, including the desired line-of-sight distance, desired incident angle, and shooting distance range; constructing a comprehensive optimization objective function within the three-dimensional safe flight space, which is a weighted combination of path length cost and safe distance cost; and using a path search or sampling planning algorithm, with the goal of minimizing the comprehensive optimization objective function, searching for the optimal path from the starting point to the ending point that passes through the observable areas near each target point and satisfies the target constraints; wherein the target constraints include at least one of the following: the line of sight from the waypoint on the path to the corresponding target point does not intersect the dynamic no-fly zone; the distance from the waypoint to the target point is within the set shooting distance range; and the angle between the line of sight and the blade surface normal satisfies the desired incident angle requirement.

[0166] Optionally, the three-dimensional safe flight space generation module is also used to: continuously read the wind turbine's operating signals at preset intervals during the UAV's inspection path execution; redetermine the maximum blade tip displacement value based on the latest operating signals within each interval, and update the three-dimensional safe flight space; the path planning module is also used to, when it is determined that the updated three-dimensional safe flight space meets the path replanning conditions, take the UAV's current position as a new starting point, replan the subsequent path based on the updated three-dimensional safe flight space, and generate new obstacle avoidance trajectory instructions; wherein, the path replanning conditions meet any of the following: the operating state crosses the working condition interval, causing the dynamic safety radius to be greater than the preset radius; the yaw angle or wind turbine azimuth angle change causes the drift caused by blade sweeping to be greater than the preset angle; when it is detected that the updated safe flight space intersects with the currently unexecuted path, the distance between the UAV's position and the boundary is less than the preset distance.

[0167] The inspection path planning device for operating wind turbines provided in this embodiment first acquires the operating signals of the wind turbine in real time during the online inspection phase. Based on at least one of the wind speed, turbulence intensity, rotational speed, and blade pitch angle acquired from the operating signals, it queries the mapping relationship pre-generated in the offline phase to generate the maximum blade tip displacement value that the flexible deformation of the blade can achieve under the current operating conditions. Based on the yaw angle and wind turbine azimuth angle acquired from the operating signals, it constructs a rigid body kinematic model of the blade and determines the basic no-fly zone formed by the rigid body sweep of the blade. Then, based on the maximum blade tip displacement value and the comprehensive error, it determines the dynamic safety radius and expands the basic no-fly zone outward with the dynamic safety radius as the expansion distance. The remaining space in the preset flight space after removing the dynamic no-fly zone is determined as the three-dimensional safe flight space. Then, under the constraint of the three-dimensional safe flight space, it plans the three-dimensional inspection path of the UAV and adaptively updates the three-dimensional safe flight space based on the real-time acquired operating signals during the UAV's flight inspection process, and adjusts the flight trajectory based on the updated three-dimensional safe flight space. In other words, an offline mapping relationship is first established between the maximum displacement value of the blade tip and the specific operating conditions. During the inspection process, the flexible deformation boundary of the blade is combined with the online acquired operating signals to dynamically generate a three-dimensional safe flight space. Under the constraints of this dynamic three-dimensional safe space, the path of the UAV to inspect the wind turbine is automatically planned. The deterministic motion and flexible deformation of the blade can be accurately described without relying on real-time visual perception. While ensuring the accuracy of the safety boundary, the requirements for online computing power in the online inspection process can be significantly reduced. This allows UAVs to perform efficient and safe automatic inspections of wind turbines in wind farms without shutting down the system. The wind turbines inspected by UAVs are more robust.

[0168] Accordingly, this disclosure also provides an inspection path planning device for operating wind turbines, the device including a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the steps as described in the above embodiments of the inspection path planning method for operating wind turbines.

[0169] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps described above in the embodiments of the inspection path planning method for operating wind turbines.

[0170] This disclosure also provides a computer device, the computer device including a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein when the computer-readable instructions are executed by the processor, they implement the various steps in the above-described embodiments of the inspection path planning method for operating wind turbines.

[0171] The specific implementation process of the functions and roles of each module in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0172] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0173] The foregoing has described specific embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0174] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention applied herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not claimed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0175] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0176] The above description is merely a preferred embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for planning inspection paths for operating wind turbines, characterized in that, The method includes: During the online inspection phase, the operating signals of the wind turbine are acquired in real time. The operating signals include at least one of the following: yaw angle, rotor azimuth angle, wind speed, turbulence intensity, rotational speed, and blade pitch angle. Based on at least one of the wind speed, turbulence intensity, rotational speed and blade pitch angle obtained from the operating signal, query the mapping relationship pre-generated in the offline stage to generate the maximum blade tip displacement value that the blade flexible deformation can achieve under the current operating conditions. Based on the yaw angle and wind turbine azimuth angle obtained from the operating signals, a rigid body kinematic model of the blade is constructed to determine the basic no-fly zone formed by the rigid body sweep of the blade. The dynamic safety radius is determined based on the maximum blade tip displacement and the comprehensive error, wherein the comprehensive error includes at least one of model error, measurement error, and engineering error. Using the dynamic safety radius as the expansion distance, the basic no-fly zone is expanded outward to obtain the dynamic no-fly zone. The remaining space after removing the dynamic no-fly zone from the preset flight space is determined as the three-dimensional safe flight space. Under the constraints of the three-dimensional safe flight space, the three-dimensional inspection path of the UAV is planned, and the three-dimensional safe flight space is adaptively updated according to the real-time acquired operation signals during the UAV flight inspection process, and the flight trajectory is adjusted based on the updated three-dimensional safe flight space.

2. The method according to claim 1, characterized in that, Before acquiring the wind turbine's operating signals in real time during the online inspection phase, the method further includes: A dynamic model incorporating the flexible deformation characteristics of blades was established based on the structural parameters of the wind turbine. Multiple operating conditions are enumerated, and the dynamic model is simulated offline based on multiple operating conditions to calculate the maximum displacement envelope that the blade can achieve after flexible deformation under each operating condition. Establish a mapping relationship between online measurable operating signals corresponding to various operating conditions and their corresponding maximum displacement envelopes, and store it as an offline mapping database.

3. The method according to claim 2, characterized in that, Calculate the maximum displacement envelope that the blade can achieve after flexible deformation under each operating condition, including: The blade is simplified into a flexible beam model distributed along the span, and the mass distribution, damping distribution, and stiffness distribution of the flexible beam model are determined. The first K natural modes and natural frequencies of the blade are solved by the modal superposition method, and the continuous deformation of the blade is transformed into a reduced-order dynamic system with modal coordinates as variables based on the solution results, where K is a positive integer; Based on the wind speed, turbulence intensity, rotational speed and blade pitch angle under each operating condition, the equivalent distributed load acting on the blade is determined; The equivalent distributed load is applied to the reduced-order dynamic system, and the time history of blade deformation is solved. The maximum displacement value at any specified position on the blade tip or blade is determined from the time history, and the maximum displacement value is used as the maximum displacement envelope under each operating condition.

4. The method according to claim 2, characterized in that, The step of querying a pre-generated mapping relationship from the offline stage based on at least one of the wind speed, turbulence intensity, rotational speed, and blade pitch angle obtained from the operating signal to generate the maximum tip displacement value that the blade can achieve under the current operating conditions includes: The offline mapping database is retrieved using at least one of the following: wind speed, turbulence intensity, rotational speed, and blade pitch angle, which are obtained in real-time online operating signals. When the real-time operating signal matches the discrete operating point in the offline mapping database, the tip displacement value in the corresponding maximum displacement envelope is determined as the tip maximum displacement value. When the real-time operating signal is between discrete operating points in the offline mapping database, the maximum blade tip displacement value under the current operating signal is determined by interpolation.

5. The method according to claim 4, characterized in that, The determination of the dynamic safety radius based on the maximum blade tip displacement and the overall error includes: The dynamic safety radius is obtained based on the maximum tip displacement, safety amplification factor, and fixed safety margin. The safety amplification factor is used to compensate for the deviation between simulation and actual measurement, and the fixed safety margin is used to cover the positioning error, control error and communication delay error of the UAV.

6. The method according to any one of claims 1 to 5, characterized in that, Based on the yaw angle and rotor azimuth angle obtained from the operating signals, a rigid body kinematic model of the blade is constructed to determine the basic no-fly zone formed by the rigid body sweep of the blade, including: Based on the three-dimensional geometric model and operating signals of the wind turbine constructed in the offline stage, a coordinate transformation chain is constructed from the ground inertial coordinate system through the nacelle yaw coordinate system rotating with the yaw angle to the wind turbine coordinate system rotating with the wind turbine azimuth angle. Based on the three-dimensional geometric model of the wind turbine, the local coordinates of any point on the blade in the wind turbine coordinate system are determined. Using the coordinate transformation chain, the local coordinates of any point on the blade are transformed sequentially to the nacelle yaw coordinate system and then to the ground inertial coordinate system to obtain the instantaneous position coordinates of the arbitrary point in the ground inertial coordinate system. Within a planned time window, the volume set formed by the rigid body sweep of the blades is determined and merged with static obstacles to obtain the basic no-fly zone formed by the rigid body sweep of the blades.

7. The method according to any one of claims 1 to 5, characterized in that, Planning a three-dimensional inspection path for the UAV under the constraints of the aforementioned three-dimensional safe flight space includes: Set the start and end points of the inspection task and the sequence of target points on the blade surface that need to be imaged, and set the imaging requirements, including the desired viewing distance, the desired incident angle and the shooting distance range. Within the three-dimensional safe flight space, a comprehensive optimization objective function is constructed, which is a weighted combination of path length cost and safe distance cost; The optimal path is searched from the starting point to the ending point, passing through the observable areas near each target point in sequence, and satisfying the target constraints, with the goal of minimizing the comprehensive optimization objective function. The target constraints include at least one of the following: the line of sight from waypoints on the path to the corresponding target point does not intersect the dynamic no-fly zone; the distance from the waypoint to the target point is within the set shooting distance range; and the angle between the line of sight and the normal to the blade surface meets the desired angle of incidence requirement.

8. The method according to any one of claims 1 to 5, characterized in that, During UAV flight inspections, the three-dimensional safe flight space is adaptively updated based on real-time acquired operational signals, and the flight trajectory is adjusted based on the updated three-dimensional safe flight space, including: During the inspection process, the drone continuously reads the operating signals of the wind turbine at preset intervals. The maximum tip displacement value is re-determined based on the latest operating signals in each cycle, and the three-dimensional safe flight space is updated. Once the updated 3D safe flight space is determined to meet the path replanning conditions, the current position of the UAV is taken as the new starting point, the subsequent path is replanned based on the updated 3D safe flight space, and a new obstacle avoidance trajectory instruction is generated. Among them, the path replanning conditions meet any of the following: the operating state crosses the working condition interval, causing the dynamic safety radius to be greater than the preset radius; the change in yaw angle or wind turbine azimuth angle causes the drift caused by blade sweeping to be greater than the preset angle; when the updated safe flight space is detected to intersect with the currently unexecuted path; and the distance between the UAV position and the boundary is less than the preset distance.

9. A path planning device for inspecting operating wind turbines, characterized in that, The inspection path planning device includes: a data acquisition module, an offline query module, a basic no-fly zone generation module, a dynamic safety radius generation module, a three-dimensional safe flight space generation module, and a path planning module; The data acquisition module is used to acquire the operating signals of the wind turbine in real time during the online inspection phase. The operating signals include at least one of the following: yaw angle, rotor azimuth angle, wind speed, turbulence intensity, rotational speed, and blade pitch angle. The offline query module is used to query the mapping relationship pre-generated in the offline stage based on at least one of the wind speed, turbulence intensity, rotational speed and blade pitch angle obtained from the operation signal, and generate the maximum blade tip displacement value that the flexible deformation of the blade can achieve under the current operating conditions. The basic no-fly zone generation module is used to construct a rigid body kinematic model of the blade based on the yaw angle and wind turbine azimuth angle obtained from the operation signal, and to determine the basic no-fly zone formed by the rigid body sweep of the blade. The dynamic safety radius generation module is used to determine the dynamic safety radius based on the maximum blade tip displacement value and the comprehensive error, wherein the comprehensive error includes at least one of model error, measurement error and engineering error. The three-dimensional safe flight space generation module is used to expand the basic no-fly zone outward with the dynamic safety radius as the expansion distance to obtain the dynamic no-fly zone, and to determine the remaining space in the preset flight space after removing the dynamic no-fly zone as the three-dimensional safe flight space. The path planning module is used to plan the three-dimensional inspection path of the UAV under the constraints of the three-dimensional safe flight space. The three-dimensional safe flight space generation module is also used to adaptively update the three-dimensional safe flight space based on the real-time acquired operating signals during the UAV flight inspection process. The path planning module is also used to adjust the flight trajectory based on the updated three-dimensional safe flight space.

10. An electronic device, characterized in that, include: processor; And a memory storing computer-readable instructions that, when executed by the processor, implement the steps of the inspection path planning method for operating wind turbines as described in any one of claims 1-8.

Images