A high-precision self-adaptive control method, device, equipment, medium and product suitable for a wind turbine nacelle mechanical arm on a shaking base

Abstract

The application discloses a wind turbine nacelle mechanical arm high-precision adaptive control method, device, equipment, medium and product suitable for a rocking base, relates to the offshore wind power technical field, and the method comprises the following steps: obtaining acceleration data, temperature data and current data; the acceleration data is subjected to frequency domain analysis and filter separation treatment, and disturbance acceleration is obtained; inertia disturbance torque corresponding to each joint is determined based on the disturbance acceleration; the inertia disturbance torque is converted into corresponding compensation current values, and the compensation current values are superimposed into the driving current of the corresponding joint to obtain the driving instruction after compensation; the actual length of the connecting rod is obtained by operating based on the temperature data, the current data, the connecting rod material parameters and the reference temperature, and the kinematics parameters of the mechanical arm are updated based on the actual length of the connecting rod to obtain the corrected kinematics model; the driving instruction of each joint is generated based on the driving instruction after compensation and the corrected kinematics model. The method can effectively improve the operation precision of the mechanical arm under the rocking working condition.

Description

Technical Field

[0001] This application relates to the field of offshore wind power technology, and in particular to a high-precision adaptive control method, device, equipment, medium and product for a wind turbine nacelle robotic arm suitable for a swaying base. Background Technology

[0002] The offshore wind power industry is developing rapidly, and automated and intelligent operation and maintenance have become the mainstream trend in wind turbine nacelle operations. The robotic arm built into the nacelle can perform a variety of operation and maintenance tasks such as bolt tightening, line inspection, and component disassembly and assembly. However, the disturbances of wind and waves at sea and the operation of the unit itself will cause the nacelle to shake continuously. This working condition greatly increases the difficulty of precise control of the robotic arm.

[0003] Currently, the mainstream robotic arm control methods all use kinematic models with fixed parameters and conventional current control logic, and the drive commands are generated entirely according to the preset program; when carrying out assembly-type contact operations, only a single position control mode is used, and the operation status is judged by a fixed contact force threshold.

[0004] The aforementioned traditional control modes are insufficient to counteract the effects of base swaying and component deformation, ultimately leading to insufficient operational accuracy of the robotic arm. Summary of the Invention

[0005] This application provides a high-precision adaptive control method, device, equipment, medium, and product for a wind turbine nacelle robotic arm with a swaying base, which can effectively improve the operating accuracy of the robotic arm under swaying conditions.

[0006] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, this application provides a high-precision adaptive control method for a wind turbine nacelle robotic arm suitable for a swaying base, comprising: Acquire acceleration data of the wind turbine nacelle, temperature data of the robotic arm linkage, and current data of the joint motor; Frequency domain analysis and filtering separation are performed on the acceleration data to obtain the interference acceleration; The inertial disturbance torque corresponding to each joint is determined based on the disturbance acceleration; the inertial disturbance torque is converted into the corresponding compensation current value, and the compensation current value is superimposed on the driving current of the corresponding joint to obtain the compensated driving command; The actual length of the link is obtained by calculation based on temperature data, current data, link material parameters and reference temperature, and the kinematic parameters of the robot arm are updated based on the actual length of the link to obtain the corrected kinematic model. Based on the compensated drive commands and the corrected kinematic model, drive commands for each joint are generated to control the movement of the robotic arm.

[0007] Among the possible implementations are: Acquire force data at the end effector of the robotic arm; During the assembly process, if the force data at the end exceeds the dynamic threshold, it is determined to be in a contact state, and the current position control mode is switched to the variable stiffness impedance control mode. Based on the nacelle vibration energy, the virtual stiffness in the variable stiffness impedance control mode is negatively correlated to achieve compliant assembly. The nacelle vibration energy is obtained by performing spectral analysis on the acceleration data.

[0008] In some possible implementations, frequency domain analysis and filtering are performed on the acceleration data to obtain the interference acceleration, including: The acceleration data is subjected to a fast Fourier transform to obtain the frequency domain amplitude spectrum; based on the frequency domain amplitude spectrum, the local maximum amplitude point is searched in the preset low-frequency and high-frequency bands to obtain the principal vibration frequency; The center frequency of the digital notch filter is updated based on the principal vibration frequency, and the principal vibration frequency component in the acceleration data is filtered out using the updated notch filter to obtain the interference acceleration.

[0009] In some possible implementations, the actual length of the connecting rod is obtained by calculation based on temperature data, current data, connecting rod material parameters, and a reference temperature, including: Calculate the difference between the temperature data and the reference temperature, and multiply the difference by the material's coefficient of thermal expansion to obtain the first correction amount; Multiply the square of the current data by the current thermal effect coefficient to obtain the second correction amount; Add the first correction amount, the second correction amount, and the benchmark proportional coefficient to obtain the comprehensive correction coefficient; Multiply the nominal length of the connecting rod by the comprehensive correction factor to obtain the actual length of the connecting rod.

[0010] In some possible implementations, the virtual stiffness in the variable stiffness impedance control mode is negatively correlated with the nacelle vibration energy to achieve compliant assembly, including: Obtain the preset virtual stiffness threshold and attenuation coefficient; The virtual stiffness threshold is multiplied by the target exponent value to obtain the real-time virtual stiffness; where the target exponent value is the product of the damping coefficient with the natural constant as the base and the exponent of the exponent value being negative and the cabin vibration energy. The real-time virtual stiffness is used as the current stiffness parameter in the variable stiffness impedance control mode.

[0011] In some possible implementations, the process of determining the dynamic threshold includes: Obtain the preset basic contact force threshold and acceleration coupling coefficient; Calculate the product of the acceleration coupling coefficient and the real-time amplitude of the acceleration data, and add this product to the basic contact force threshold to obtain the dynamic threshold.

[0012] Secondly, this application provides a high-precision adaptive control device for a wind turbine nacelle robotic arm suitable for a swaying base, comprising: The acquisition module is used to acquire acceleration data of the wind turbine nacelle, temperature data of the robotic arm linkage, and current data of the joint motor. The calculation and processing module is used to perform frequency domain analysis and filtering separation on the acceleration data to obtain the disturbance acceleration; determine the inertial disturbance torque corresponding to each joint based on the disturbance acceleration; convert the inertial disturbance torque into the corresponding compensation current value, and superimpose the compensation current value into the driving current of the corresponding joint to obtain the compensated driving command; perform calculations based on temperature data, current data, link material parameters and reference temperature to obtain the actual length of the link, and update the kinematic parameters of the robot arm based on the actual length of the link to obtain the corrected kinematic model; The control module is used to generate drive commands for each joint based on the compensated drive commands and the corrected kinematic model, so as to control the movement of the robotic arm.

[0013] Thirdly, this application provides a computing device, including a memory and a processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of the first aspects.

[0014] Fourthly, this application provides a computer-readable storage medium for storing a computer program for performing the method as described in any one of the first aspects.

[0015] Fifthly, this application provides a computer program product comprising one or more computer instructions, wherein when the computer instructions are executed by a computer, the computer performs the method as described in any one of the first aspects.

[0016] As can be seen from the above technical solution, this application has at least the following beneficial effects: In this application, by acquiring the acceleration of the wind turbine nacelle, the temperature of the robotic arm link, and the joint motor current, disturbance acceleration is extracted through frequency domain analysis and filtering. This disturbance acceleration is then used to perform joint torque compensation and generate compensated drive commands. Simultaneously, the actual length of the link is calculated based on temperature, current, and related parameters, updating the corrected kinematic model. Finally, the compensated drive commands and the corrected model are combined to output joint drive commands, achieving robotic arm motion control. This scheme addresses the inertial disturbances caused by nacelle swaying and the link size changes caused by temperature and motor heating, simultaneously optimizing the robotic arm control logic from both the drive command and motion model dimensions. This overcomes the shortcomings of traditional schemes that do not address working condition disturbances and component deformation. Therefore, it improves the deficiencies of traditional control methods and effectively enhances the operational accuracy of the robotic arm under swaying conditions.

[0017] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0018] Figure 1 An application environment diagram for a high-precision adaptive control method for a wind turbine nacelle robotic arm with a swaying base, provided in an embodiment of this application; Figure 2 A flowchart illustrating a high-precision adaptive control method for a wind turbine nacelle robotic arm with a swaying base, provided as an embodiment of this application; Figure 3 A structural diagram of a high-precision adaptive control device for a wind turbine nacelle robotic arm suitable for a swaying base, provided in an embodiment of this application; Figure 4 This is a schematic diagram of a computing device provided in an embodiment of this application. Detailed Implementation

[0019] The terms "first," "second," and "third," etc., used in this application specification and accompanying drawings are used to distinguish different objects, not to limit a specific order.

[0020] In the embodiments of the present application, words such as "exemplary" or "for example" are used to serve as an example, illustration or description. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application shall not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete manner.

[0021] For a clear and concise description of the following embodiments, a brief introduction to the related art is provided first: A wind turbine nacelle is a closed cabin that accommodates components such as the gearbox and generator of a wind turbine, and is also the mounting base for the nacelle operation and maintenance manipulator. Affected by the impact of offshore wind and waves and the self-operation of the unit, the wind turbine nacelle will continuously generate swing and vibration, forming a swaying base working condition. The manipulator connecting rod is the rigid load-bearing structure of the manipulator, and its geometric length will change with the ambient temperature and the heat generated by the operation of adjacent motors. As the power component of the manipulator, the joint motor operates by relying on the input driving current. The swaying of the nacelle will generate additional inertial disturbance torque at each joint of the manipulator, which interferes with the normal driving state of the motor. A six-dimensional inertial measurement unit, a temperature sensor, a current acquisition device and a six-dimensional force sensor are the supporting data acquisition components of the present solution, which can respectively obtain the nacelle acceleration, connecting rod temperature, motor current and force information at the end of the manipulator in real time.

[0022] In existing control methods, manipulators generally adopt fixed-parameter kinematic models and conventional current control modes. They do not adjust driving signals in combination with the inertial disturbance caused by nacelle vibration, nor do they consider the change of connecting rod size caused by temperature and motor heat generation, so there is a deviation between the motion model and the actual state of the manipulator. The superposition of the two types of problems above makes it difficult for traditional control methods to adapt to the complex operating environment of the swaying nacelle base, and ultimately results in poor operation performance of the manipulator.

[0023] In view of this, embodiments of the present application provide a high-precision adaptive control method for wind turbine nacelle manipulators suitable for swaying bases. To make the technical solution of the present application clearer and easier to understand, the application scenario of the technical solution of the present application is introduced below with reference to the accompanying drawings. As Figure 1 shown, the figure is a schematic diagram of an application environment provided by an embodiment of the present application.

[0024] In this application environment, the main control unit forms linkage interaction with various data acquisition devices and manipulator joint motors. The main control unit issues acquisition instructions to the six-dimensional inertial measurement unit, the temperature sensor, the current acquisition device and the six-dimensional force sensor, and obtains the wind turbine nacelle acceleration, manipulator connecting rod temperature, joint motor current and end force data of the manipulator in real time; performs arithmetic processing on the collected data, completes operations such as interference extraction, torque compensation and kinematic model correction, and issues the finally generated joint driving instructions to each joint motor to drive the manipulator to complete corresponding operation and maintenance actions.

[0025] To make the technical solution of this application clearer and easier to understand, the following describes a high-precision adaptive control method for a wind turbine nacelle robotic arm suitable for a swaying base, based on the above application scenarios and with the main control unit as the execution subject. For example... Figure 2 As shown in the figure, this is a flowchart illustrating a high-precision adaptive control method for a wind turbine nacelle robotic arm suitable for a swaying base, provided in an embodiment of this application. The high-precision adaptive control method for a wind turbine nacelle robotic arm suitable for a swaying base includes: S201. Acquire acceleration data of the wind turbine nacelle, temperature data of the robotic arm linkage, and current data of the joint motor.

[0026] The wind turbine nacelle is a sealed enclosure on top of the wind turbine that houses equipment such as the gearbox and generator. In this solution, the maintenance robotic arm is fixedly installed inside the nacelle. When the wind turbine is running, the nacelle will experience low-frequency swaying and high-frequency vibration, and the robotic arm mounted inside the nacelle will also sway accordingly.

[0027] Acceleration data can be collected in real time by a six-dimensional inertial measurement unit at the base of the robotic arm, which can reflect the spatial motion of the wind turbine nacelle and cover the motion information corresponding to tower sway and gearbox vibration.

[0028] The robotic arm linkages are the rigid components that make up the wind power operation and maintenance robotic arm. The robotic arm relies on various linkages to complete actions such as extension, displacement, and attitude adjustment. Temperature changes inside the nacelle, heat dissipation from surrounding equipment, and heat generated by the motor will all cause the linkages to expand and contract due to temperature changes.

[0029] Temperature data can be collected by temperature sensors embedded inside the robotic arm's linkage, reflecting the linkage's current temperature. Fluctuations in cabin temperature throughout the day and night, as well as continuous heat dissipation from surrounding equipment, will alter the linkage's temperature. In low-temperature environments and after prolonged operation, the sensors will collect different temperature readings.

[0030] Joint motors are installed at various joints of the robotic arm. The operation of the motors drives the linkages to rotate, thereby changing the posture of the robotic arm. During operation, the motors generate heat, which alters the shape of the surrounding linkages, resulting in lifting, rotating, and other similar movements.

[0031] The current data refers to the real-time operating current value collected during the operation of the joint motor, which can reflect the motor's operating status and heat generation. When the range of motion or operating speed of the robotic arm changes, the motor's operating current will also change accordingly.

[0032] For example, during the operation of the robotic arm, the main control unit uniformly issues data acquisition commands and simultaneously activates various sensing devices to carry out data acquisition. The main control unit activates the six-dimensional inertial measurement unit installed on the robotic arm base and continuously receives the real-time acceleration data of the wind turbine nacelle output by the six-dimensional inertial measurement unit. The main control unit activates the temperature sensor embedded inside the robotic arm link and continuously reads the temperature data of the robotic arm link transmitted back by the temperature sensor. At the same time, the main control unit activates the current acquisition device corresponding to the joint motor and acquires the joint motor current data monitored by the current acquisition device in real time.

[0033] S202. Perform frequency domain analysis and filtering separation on the acceleration data to obtain the interference acceleration.

[0034] One possible approach is to perform a fast Fourier transform on the acceleration data to obtain the frequency domain amplitude spectrum; based on the frequency domain amplitude spectrum, search for local maximum amplitude points in preset low-frequency and high-frequency bands to obtain the principal vibration frequency; update the center frequency of the digital notch filter according to the principal vibration frequency, and use the updated notch filter to filter out the principal vibration frequency component in the acceleration data to obtain the interference acceleration.

[0035] Frequency domain analysis is a signal processing method for continuously changing acceleration data. It can convert time-domain signals that change over time into signals with frequency as the dimension, thereby distinguishing the different frequency components contained in the data.

[0036] Filtering and separation processing is an operation that uses specialized filtering equipment to remove specified frequency components from a signal. It can distinguish between effective signals and interference signals and extract the target signal content.

[0037] Disturbance acceleration refers to the acceleration signal generated by the swaying of the wind turbine nacelle and the vibration of the equipment. This type of signal can affect the normal operation of the robotic arm and is also the target of torque compensation in this solution.

[0038] Fast Fourier Transform (FFT) is a commonly used signal processing method that can directly convert acceleration data in the time domain into a spectrum in the frequency domain, visually displaying the signal amplitude corresponding to different frequencies.

[0039] The frequency domain amplitude spectrum is a graph formed after acceleration data has been converted. The horizontal axis of the graph represents the signal frequency, and the vertical axis represents the signal amplitude at the corresponding frequency, which can intuitively present the frequency distribution of the signal.

[0040] The low-frequency band is a pre-defined frequency range that corresponds to the frequency range of signals generated by the overall oscillation of the wind turbine nacelle.

[0041] The high-frequency band is a pre-defined frequency range that corresponds to the frequency range of signals generated by the operation of the gearbox inside the engine room.

[0042] The dominant vibration frequency is the frequency point with the largest signal amplitude in the low-frequency and high-frequency bands, representing the dominant frequency of the current vibration in the cabin.

[0043] A digital notch filter is a functional module used for signal filtering, which can selectively filter out signal components of a specific single frequency.

[0044] The center frequency is the currently set operating frequency of the digital notch filter, and the filter will perform signal filtering operations around this frequency.

[0045] For example, the main control unit retrieves the collected wind turbine nacelle acceleration data, performs a Fast Fourier Transform (FFT) operation on the data, and generates the corresponding frequency domain amplitude spectrum. Referring to pre-defined low-frequency and high-frequency bands, the main control unit finds the position with the largest amplitude value within each frequency range and determines the corresponding frequency as the dominant vibration frequency. Using the identified dominant vibration frequency, the main control unit resets the center frequency of the digital notch filter. After the parameter update, the main control unit runs the digital notch filter to remove signal components belonging to the dominant vibration frequency from the acceleration data, ultimately obtaining the interference acceleration.

[0046] S203. Determine the inertial disturbance torque corresponding to each joint based on the disturbance acceleration; convert the inertial disturbance torque into the corresponding compensation current value, and superimpose the compensation current value into the driving current of the corresponding joint to obtain the compensated driving command.

[0047] Inertial disturbance torque is the torque generated at various joint positions when the cabin vibration drives the movement of the robotic arm. This torque will change the original operating state of the robotic arm.

[0048] The compensation current value is the additional current increment given in the motor current loop that needs to be added to counteract the inertial disturbance torque.

[0049] Drive current is a current signal input to the joint motor to control the motor's operation. The motor relies on this type of current to complete its rotation.

[0050] The compensated drive command is a new control command formed by superimposing the torque correction signal, which is used to control the normal operation of the joint motor.

[0051] For example, the main control unit retrieves the processed disturbance acceleration, solves for the inertial disturbance torque joint by joint based on the Newton-Euler dynamics model, reads the current angle and angular velocity data of each joint of the robotic arm in real time, constructs the base entrainment motion transmission matrix in combination with the base disturbance acceleration, calculates the entrainment acceleration of the center of mass of each link layer by layer in a forward recursive manner, solves for the inertial force induced by the entrainment acceleration in combination with the mass of each link, and then recursively calculates along the links in a reverse manner, taking the moment of the inertial force towards the corresponding joint rotation axis and accumulating them, and respectively calculates the inertial disturbance torque corresponding to each of the six joints. The main control unit performs the conversion from torque to compensation current and retrieves the factory-calibrated torque constants of each joint motor stored locally. The electromagnetic torque of the motor has a linear relationship with the q-axis current, and the compensation current increment... The calculation process is as follows: The compensation current increment ΔI for each joint is calculated one by one using the above formula. At the same time, current limiting logic is executed to set upper and lower current thresholds to prevent the compensation current from exceeding the limit and triggering the electromagnetic torque of the motor when the nacelle shakes violently. Finally, the main control unit synthesizes the compensation drive command, first obtaining the basic original drive current of each joint through internal position loop, velocity loop and conventional dynamics feedforward calculation. The calculation process for the compensated drive current command is as follows: in, This is the compensated drive current.

[0052] The main control unit will send the compensated drive command calculated according to the formula to the corresponding joint driver. The motor will synchronously output the basic torque for trajectory tracking and the reverse compensation torque to counteract the sway of the base, so as to realize the real-time feedforward suppression of the vibration inertial disturbance of the base.

[0053] S204. Based on temperature data, current data, link material parameters, and reference temperature, calculations are performed to obtain the actual length of the link. The kinematic parameters of the robotic arm are then updated based on the actual length of the link to obtain the corrected kinematic model.

[0054] The connecting rod material parameters include the material's coefficient of thermal expansion and the coefficient of thermal effect of electric current.

[0055] The process of determining the actual length of the connecting rod can be as follows: calculate the difference between the temperature data and the reference temperature, and multiply the difference by the coefficient of thermal expansion of the material to obtain the first correction amount; multiply the square value of the current data by the current thermal effect coefficient to obtain the second correction amount; add the first correction amount, the second correction amount and the reference proportional coefficient to obtain the comprehensive correction coefficient; multiply the nominal length of the connecting rod by the comprehensive correction coefficient to obtain the actual length of the connecting rod.

[0056] Link material parameters are a set of parameters that describe the physical properties of the robotic arm link itself. This solution includes two types of fixed parameters used in calculations related to the link length.

[0057] The coefficient of thermal expansion is a fixed parameter corresponding to the material of the connecting rod, which can reflect the influence of temperature changes on the dimensions of the connecting rod.

[0058] The current thermal effect coefficient is a fixed parameter set in conjunction with the operating conditions of the connecting rod and joint motor, which reflects the impact of the heat generated by the motor on the dimensions of the connecting rod.

[0059] The reference temperature is a pre-set standard temperature value, which serves as a comparison benchmark to measure the current temperature change of the connecting rod.

[0060] The first correction is a value calculated by combining temperature changes and material properties, used to reflect the change in the length of the connecting rod caused by temperature factors.

[0061] The second correction is a value calculated by combining the motor's operating current with the corresponding coefficient, used to reflect the change in the connecting rod length caused by the motor's heat generation factor.

[0062] The reference ratio coefficient is a fixed value, representing the length ratio of the link under standard conditions, and is the basic value for length calculation.

[0063] The comprehensive correction factor is a value obtained by integrating multiple influencing factors, which can uniformly reflect the overall changes in the length of the connecting rod caused by temperature and motor heat generation.

[0064] The nominal length of the link is the original length value of the robotic arm link under standard working conditions.

[0065] The actual length of the link is a real-time value calculated by taking into account various influencing factors.

[0066] Kinematic parameters of a robotic arm are parameters that describe the relationship between the structure, position, and movement of the robotic arm, and link length is one such parameter.

[0067] The revised kinematic model is a completely new model formed after updating the structural parameters, used to describe the current motion and positional relationship of the robotic arm.

[0068] For example, the main control unit calculates the actual length of the connecting rod based on the rod's temperature data, current data, material parameters, and reference temperature, using the following formula:

[0069] in, This is the actual length of the connecting rod; It is the nominal length of the connecting rod, that is, the design length under standard conditions; The coefficient of thermal expansion of the material; The temperature data of the connecting rod, i.e. the temperature of the connecting rod body, is collected in real time by a temperature sensor embedded inside the connecting rod. This is a reference temperature, for example, 20℃ (standard ambient temperature); The coefficient of thermal effect of electric current; The root mean square value of the joint motor current is obtained by the main control unit performing root mean square calculation on the raw current data sampled in real time from the motor, reflecting the current degree of heat generation of the motor.

[0070] Within each interpolation cycle (e.g., 4ms), the control unit dynamically updates the link length parameters of the robotic arm based on the calculated real-time actual link length. The main control unit then performs a forward kinematics model update calculation based on the updated link length parameters to ensure that the theoretical spatial position of the robotic arm's end effector calculated by the motion controller remains consistent with the actual physical spatial position of the robotic arm. Simultaneously, the main control unit recalculates the target rotation angles corresponding to each joint of the robotic arm based on the updated link length parameters, thereby eliminating the end effector positioning error caused by link thermal deformation.

[0071] S205. Based on the compensated drive commands and the corrected kinematic model, drive commands for each joint are generated to control the movement of the robotic arm.

[0072] The compensated drive command is a control signal formed by superimposing the inertial disturbance torque signal, which is specifically applied to the joint motors of the robotic arm.

[0073] The revised kinematic model is a model formed after updating the relevant parameters of the link length, and is used to match the current structural state of the robotic arm.

[0074] The joint drive command is the final control signal sent to the joint motor, and the motor completes the operation according to the signal.

[0075] For example, the main control unit calls the compensated drive commands and the corrected kinematic model, combines the two types of data for calculation, and generates drive commands for each joint. The main control unit then sends the generated drive commands to the motors of each joint, thereby controlling the robotic arm to complete the corresponding movements.

[0076] Based on the above, a high-precision adaptive control method for a wind turbine nacelle robotic arm with a swaying base is proposed. This method acquires data on wind turbine nacelle acceleration, robotic arm link temperature, joint motor current, and mechanical end-effector force. Through frequency domain analysis and filtering, disturbance acceleration is extracted. Combined with the disturbance acceleration, joint torque compensation is performed, and compensated drive commands are generated. Simultaneously, the actual link length is calculated using temperature, current, and related parameters, updating the corrected kinematic model. Finally, the compensated drive commands and the corrected model are combined to output joint drive commands, achieving robotic arm motion control. This scheme addresses the inertial disturbances caused by nacelle swaying and the link size changes caused by temperature and motor heating. It simultaneously optimizes the robotic arm control logic from both the drive command and motion model dimensions, overcoming the shortcomings of traditional methods that do not address operating condition disturbances and component deformation. This improves the deficiencies of traditional control methods and effectively enhances the operational accuracy of the robotic arm under swaying conditions.

[0077] Based on the above embodiments, this embodiment provides a detailed explanation of the process for achieving compliant assembly, specifically including: Acquire force data at the end of the robotic arm; during the assembly process, if the force data at the end exceeds the dynamic threshold, it is determined to be in a contact state, and the current position control mode is switched to the variable stiffness impedance control mode; based on the cabin vibration energy, the virtual stiffness in the variable stiffness impedance control mode is negatively correlated to achieve compliant assembly.

[0078] The end effector of the robotic arm is its outermost working structure, through which the arm contacts wind turbine components such as bolts and aviation connectors. Force sensors are installed at the end effector. Force data from the end effector is collected by a six-dimensional force / torque sensor, which characterizes the external contact forces acting on the arm's actuator. When the end effector touches various wind turbine components, the six-dimensional force / torque sensor collects the corresponding force data. The main control unit can continuously collect and receive the force data from the end effector by activating the six-dimensional force / torque sensor.

[0079] Assembly tasks refer to various precision contact operations performed by the robotic arm inside the wind turbine nacelle, such as tightening wind turbine bolts, plugging and unplugging aviation connectors, and disassembling and assembling small components. The dynamic threshold is a contact force judgment threshold value that changes in real time with the nacelle vibration state, used to accurately identify the contact state under nacelle swaying conditions. A contact state is the operational state in which the robotic arm's end effector makes substantial physical contact with the workpiece to be assembled. For example, the robotic arm gripper contacting the surface of a wind turbine bolt, or the connector end aligning and contacting the socket, are both considered contact states. Position control mode is the conventional basic control mode for the robotic arm. Using preset coordinates and trajectories as targets, it strictly controls the robotic arm's movement position and attitude. Its main function is to ensure the robotic arm's operational accuracy. It lacks flexible buffering capabilities and is used during non-contact preparatory operations such as unloaded movement of the robotic arm and workpiece alignment.

[0080] The variable stiffness impedance control mode is a flexible control mode designed for contact-type operations. By dynamically adjusting the internal virtual stiffness parameters, the robotic arm has the ability to buffer and adapt to external disturbances, effectively avoiding damage to components caused by rigid contact. For example, when the nacelle shakes significantly, this mode allows the end of the robotic arm to float slightly flexibly, dynamically changing position to follow the workpiece to be assembled, avoiding rigid pressure and impact damage to wind turbine components.

[0081] Cabin vibration energy is an energy characteristic value calculated after performing spectral analysis on the time-domain signal of cabin acceleration. It is used to quantify the severity of the overall cabin vibration. Spectral analysis is a signal processing method that performs a Fast Fourier Transform on the time-domain cabin acceleration signal to convert the time-domain signal into a frequency-domain spectrum, extracts the corresponding frequencies and amplitudes, and then further calculates the vibration energy. Negative correlation adjustment refers to the fact that the changing trends of cabin vibration energy and virtual stiffness are completely opposite. For example, the greater the cabin vibration energy, the smaller the virtual stiffness; and the smaller the vibration energy, the greater the virtual stiffness. Virtual stiffness is an important parameter in the impedance control model, used to simulate the working stiffness of the robotic arm end effector. The larger the virtual stiffness value, the stronger the rigidity of the robotic arm operation and the less prone to flexible deformation; the smaller the virtual stiffness value, the stronger the flexibility of the robotic arm operation and the easier it is to adapt to external force disturbances.

[0082] The dynamic threshold is a crucial parameter for contact state identification and can be updated in real time based on nacelle vibration conditions. The process of determining the dynamic threshold includes: Obtain the preset basic contact force threshold and acceleration coupling coefficient; calculate the product of the acceleration coupling coefficient and the real-time amplitude of the acceleration data, and add the product to the basic contact force threshold to obtain the dynamic threshold.

[0083] The real-time amplitude of the acceleration data is the instantaneous amplitude of the nacelle acceleration signal acquired by the six-dimensional inertial measurement unit at the current moment. The basic contact force threshold is a preset fixed force value, which serves as the reference threshold for determining workpiece contact when the nacelle is completely stationary. The acceleration coupling coefficient is a preset proportionality coefficient used to quantify the degree of influence of the nacelle acceleration amplitude on the contact force threshold.

[0084] For example, during assembly operations by the robotic arm of a wind turbine nacelle, the main control unit first collects real-time acceleration data of the nacelle and force data at the end of the robotic arm. Combined with preset parameters, it calculates a dynamic threshold in real time to achieve real-time identification of the robotic arm's operational status. The formula for calculating the dynamic threshold is:

[0085] in, Represents the dynamic threshold, which is the real-time criterion for determining whether the end effector of the robotic arm is in contact with the workpiece; The basic contact force threshold is a preset fixed contact judgment benchmark value when the cabin is in a stationary state; The acceleration coupling coefficient is used to characterize the weight of the influence of cabin acceleration on the contact determination threshold. It is the real-time amplitude of the cabin acceleration data, reflecting the current instantaneous sway intensity of the cabin.

[0086] The main control unit compares the collected end-effector force data with the calculated dynamic threshold. When the end-effector force data exceeds the dynamic threshold, the robot arm is determined to have entered a contact state. Immediately, the robot arm's original position control mode switches to variable stiffness impedance control mode, and the compliant assembly control logic is activated. To achieve adaptive compliant assembly under cabin sway conditions, a target impedance model matching the operating conditions needs to be constructed under the variable stiffness impedance control mode. The contact compliance characteristics of the robot arm's end-effector are dynamically adjusted through model parameters. For example, a second-order dynamic impedance model can be used to establish the correspondence between the robot arm's end-effector position deviation and the external contact force. The specific construction method of the target impedance model is as follows:

[0087] in, The virtual mass matrix determines the inertial characteristics of the robotic arm's end effector. This is a virtual damping matrix used to dissipate contact oscillation energy; This is a virtual stiffness matrix that determines the compliance of the end effector with external contact forces. The external contact force at the end is collected by the end force sensor. The desired position is the planned target position at the end of the robotic arm. This refers to the actual position, that is, the current actual position of the robotic arm's end effector. For the desired speed; This refers to the actual speed; For the desired acceleration; This is the actual acceleration.

[0088] The target impedance model accurately describes the second-order dynamic correlation between the position deviation, velocity deviation, acceleration deviation of the robotic arm end effector and the external contact force. Under the premise that the virtual mass matrix and virtual damping matrix are fixed, the compliance of the robotic arm end effector with the external contact force can be changed in real time by simply adjusting the value of the virtual stiffness matrix, so as to adapt to the cabin vibration conditions of different intensities.

[0089] To avoid the problem of fixed stiffness failing to adapt to the dynamic vibration conditions of the nacelle, instead of using fixed virtual stiffness parameters, an adaptive virtual stiffness adjustment function that is negatively correlated with the real-time vibration energy of the nacelle can be constructed. This function updates the virtual stiffness parameters in the impedance model in real time based on the nacelle vibration energy. The nacelle vibration energy is obtained by the main control unit through spectral analysis of the real-time acquired nacelle acceleration data, which can accurately quantify the severity of the current overall vibration of the nacelle.

[0090] The specific process for adjusting virtual stiffness and determining the current stiffness parameters of the variable stiffness impedance control mode based on the negative correlation of nacelle vibration energy is as follows: Obtain the preset virtual stiffness threshold and attenuation coefficient; multiply the virtual stiffness threshold by the target exponent value to obtain the real-time virtual stiffness; wherein, the target exponent value is the product of the attenuation coefficient with the natural constant as the base and the exponent of the exponent value being negative and the nacelle vibration energy; use the real-time virtual stiffness as the current stiffness parameter of the variable stiffness impedance control mode.

[0091] The exponential calculation formula is constructed with the natural constant as the base. The preset attenuation coefficient is multiplied by the real-time vibration energy of the cabin and the negative value is taken to obtain the target exponential value of negative exponential attenuation. The virtual stiffness threshold is multiplied by the target exponential value to calculate the real-time virtual stiffness at the current moment. The calculated real-time virtual stiffness is assigned to the virtual stiffness matrix of the target impedance model as the current stiffness parameter of the variable stiffness impedance control mode, so as to realize the adaptive adjustment of compliance characteristics.

[0092] The calculation process for real-time virtual stiffness is as follows:

[0093] in, The real-time virtual stiffness at the current moment determines the flexibility of the robotic arm's end effector; It is a preset virtual stiffness threshold, that is, the maximum virtual stiffness, which corresponds to the standard operating stiffness of the engine room under stable conditions. It is a natural constant; It is a preset attenuation coefficient used to control how quickly the virtual stiffness decays with vibration energy; This represents the real-time vibration energy of the cabin.

[0094] As can be seen from the formula's logic, the greater the cabin vibration energy (i.e., the more violent the cabin shaking), the larger the exponential part... The smaller the value, the lower the calculated real-time virtual stiffness, resulting in enhanced robotic arm flexibility that can adapt to cabin swaying and avoid rigid collisions. Conversely, the lower the cabin vibration energy, the closer the real-time virtual stiffness is to its maximum value, increasing robotic arm rigidity and ensuring assembly precision. The system uses the calculated real-time virtual stiffness as the operating parameter for the variable stiffness impedance control mode, driving the robotic arm to complete compliant assembly operations under the swaying base.

[0095] This embodiment uses dynamic thresholds to determine the contact state of the robotic arm and adjusts the judgment criteria in real time based on the cabin acceleration, effectively distinguishing between vibration inertial force and actual contact force to avoid misjudgment. It automatically switches between position control mode and variable stiffness impedance control mode to ensure operational accuracy during alignment and activate flexible control during contact assembly. Simultaneously, it dynamically adjusts the virtual stiffness based on the cabin vibration energy; the stronger the vibration, the lower the stiffness, and the greater the flexibility of the robotic arm, adapting to base sway and buffering impacts. Stiffness returns to normal after the cabin stabilizes, ensuring operational accuracy. The entire control logic adapts to the continuous swaying of the cabin, preventing damage to components from rigid collisions and improving assembly success rate and system environmental adaptability.

[0096] The above text combined Figures 1 to 2 This application provides a detailed description of a high-precision adaptive control method for a wind turbine nacelle robotic arm suitable for a swaying base, as provided in the embodiments of this application. The apparatus and equipment provided in the embodiments of this application will be described below with reference to the accompanying drawings.

[0097] This application also provides a high-precision adaptive control device for a wind turbine nacelle robotic arm suitable for a swaying base, such as... Figure 3 As shown in the figure, this is a structural diagram of a high-precision adaptive control device for a wind turbine nacelle robotic arm suitable for a swaying base, provided in an embodiment of this application. The device includes: The acquisition module 301 is used to acquire acceleration data of the wind turbine nacelle, temperature data of the robotic arm linkage, and current data of the joint motor. The calculation and processing module 302 is used to perform frequency domain analysis and filtering separation on the acceleration data to obtain the disturbance acceleration; determine the inertial disturbance torque corresponding to each joint based on the disturbance acceleration; convert the inertial disturbance torque into the corresponding compensation current value, and superimpose the compensation current value into the driving current of the corresponding joint to obtain the compensated driving command; perform calculations based on temperature data, current data, link material parameters and reference temperature to obtain the actual length of the link, and update the kinematic parameters of the robot arm based on the actual length of the link to obtain the corrected kinematic model. The control module 303 is used to generate drive commands for each joint based on the compensated drive commands and the corrected kinematic model, so as to control the movement of the robotic arm.

[0098] Among some possible implementations, the high-precision adaptive control device for the wind turbine nacelle robotic arm with a swaying base also includes: The compliant assembly module is used to acquire force data at the end of the robotic arm. During the assembly process, if the force data at the end exceeds the dynamic threshold, it is determined to be in a contact state, and the current position control mode is switched to the variable stiffness impedance control mode. Based on the cabin vibration energy, the virtual stiffness in the variable stiffness impedance control mode is negatively correlated to achieve compliant assembly. The cabin vibration energy is obtained by performing spectrum analysis on the acceleration data.

[0099] In some possible implementations, the computation processing module 302 specifically includes: The acceleration data is subjected to a fast Fourier transform to obtain the frequency domain amplitude spectrum. Based on the frequency domain amplitude spectrum, the local maximum amplitude point is searched in the preset low-frequency and high-frequency bands to obtain the principal vibration frequency. The center frequency of the digital notch filter is updated according to the principal vibration frequency, and the principal vibration frequency component in the acceleration data is filtered out by the updated notch filter to obtain the interference acceleration.

[0100] In some possible implementations, the computation processing module 302 specifically includes: Calculate the difference between the temperature data and the reference temperature, and multiply the difference by the material's thermal expansion coefficient to obtain the first correction amount; multiply the square of the current data by the current thermal effect coefficient to obtain the second correction amount; add the first correction amount, the second correction amount, and the reference proportionality coefficient to obtain the comprehensive correction coefficient; multiply the nominal length of the connecting rod by the comprehensive correction coefficient to obtain the actual length of the connecting rod.

[0101] In some possible implementations, the compliant assembly module specifically includes: Obtain the preset virtual stiffness threshold and attenuation coefficient; multiply the virtual stiffness threshold by the target exponent value to obtain the real-time virtual stiffness; wherein, the target exponent value is the product of the attenuation coefficient with the natural constant as the base and the exponent of the exponent value being negative and the nacelle vibration energy; use the real-time virtual stiffness as the current stiffness parameter of the variable stiffness impedance control mode.

[0102] In some possible implementations, the compliant assembly module specifically includes: Obtain the preset basic contact force threshold and acceleration coupling coefficient; calculate the product of the acceleration coupling coefficient and the real-time amplitude of the acceleration data, and add the product to the basic contact force threshold to obtain the dynamic threshold.

[0103] The high-precision adaptive control device for a wind turbine nacelle robotic arm with a swaying base according to the embodiments of this application can correspondingly execute the method described in the embodiments of this application, and the other operations and / or functions of each module / unit of the high-precision adaptive control device for a wind turbine nacelle robotic arm with a swaying base are respectively for realizing Figure 2 For the sake of brevity, the corresponding processes of each method in the illustrated embodiments will not be described in detail here.

[0104] This application also provides a computing device. For example... Figure 4 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 400 includes a bus 401, a processor 402, a communication interface 403, and a memory 404. The processor 402, the memory 404, and the communication interface 403 communicate with each other via the bus 401.

[0105] Bus 401 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0106] Processor 402 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0107] Communication interface 403 is used for communication with external devices.

[0108] Memory 404 may include volatile memory, such as random access memory (RAM). Memory 404 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0109] The memory 404 stores executable code, and the processor 402 executes the executable code to perform the aforementioned high-precision adaptive control method for a wind turbine nacelle robotic arm applicable to a swaying base.

[0110] Specifically, in achieving Figure 3 In the case of the illustrated embodiment, and Figure 3When the modules or units of the high-precision adaptive control device for the wind turbine nacelle robotic arm on a swaying base described in the embodiments are implemented through software, the execution... Figure 3 The software or program code required for the functions of each module / unit can be partially or entirely stored in memory 404. Processor 402 executes the program code corresponding to each unit stored in memory 404, and executes the aforementioned high-precision adaptive control method for the wind turbine nacelle robotic arm applicable to swaying bases.

[0111] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium capable of being stored by a computing device, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the aforementioned high-precision adaptive control method for a wind turbine nacelle robotic arm with a wobbling base.

[0112] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0113] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0114] When the computer program product is executed by a computer, the computer performs any of the aforementioned methods of the high-precision adaptive control method for the manipulator arm of a swaying wind turbine nacelle. The computer program product can be a software installation package; when any of the aforementioned methods of the high-precision adaptive control method for the manipulator arm of a swaying wind turbine nacelle needs to be used, the computer program product can be downloaded and executed on the computer.

[0115] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0116] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A high-precision adaptive control method for a wind turbine nacelle robotic arm with a swaying base, characterized in that, The method includes: Acquire acceleration data of the wind turbine nacelle, temperature data of the robotic arm linkage, and current data of the joint motor; Frequency domain analysis and filtering separation are performed on the acceleration data to obtain the interference acceleration; The inertial disturbance torque corresponding to each joint is determined based on the disturbance acceleration; the inertial disturbance torque is converted into the corresponding compensation current value, and the compensation current value is superimposed on the driving current of the corresponding joint to obtain the compensated driving command; The actual length of the link is obtained by calculation based on temperature data, current data, link material parameters and reference temperature, and the kinematic parameters of the robot arm are updated based on the actual length of the link to obtain the corrected kinematic model. Based on the compensated drive commands and the corrected kinematic model, drive commands for each joint are generated to control the movement of the robotic arm.

2. The method according to claim 1, characterized in that, The method further includes: Acquire force data at the end effector of the robotic arm; During the assembly process, if the force data at the end exceeds the dynamic threshold, it is determined to be in contact state, and the current position control mode is switched to the variable stiffness impedance control mode. Based on the nacelle vibration energy, the virtual stiffness in the variable stiffness impedance control mode is negatively correlated to achieve compliant assembly; whereby the nacelle vibration energy is obtained by performing spectral analysis on the acceleration data.

3. The method according to claim 1, characterized in that, The process of performing frequency domain analysis and filtering separation on the acceleration data to obtain the interference acceleration includes: The frequency domain amplitude spectrum is obtained by performing a fast Fourier transform on the acceleration data; Based on the frequency domain amplitude spectrum, the local maximum amplitude point is searched in the preset low-frequency and high-frequency bands to obtain the main vibration frequency; The center frequency of the digital notch filter is updated based on the principal vibration frequency, and the principal vibration frequency component in the acceleration data is filtered out using the updated notch filter to obtain the interference acceleration.

4. The method according to claim 1, characterized in that, The connecting rod material parameters include the coefficient of thermal expansion and the coefficient of thermal effect of current. The calculation based on temperature data, current data, connecting rod material parameters, and reference temperature to obtain the actual length of the connecting rod includes: Calculate the difference between the temperature data and the reference temperature, and multiply the difference by the material's coefficient of thermal expansion to obtain the first correction amount; Multiply the square of the current data by the current thermal effect coefficient to obtain the second correction amount; Add the first correction amount, the second correction amount, and the benchmark proportional coefficient to obtain the comprehensive correction coefficient; Multiply the nominal length of the connecting rod by the comprehensive correction factor to obtain the actual length of the connecting rod.

5. The method according to claim 2, characterized in that, The method of negatively adjusting the virtual stiffness in the variable stiffness impedance control mode based on the nacelle vibration energy to achieve compliant assembly includes: Obtain the preset virtual stiffness threshold and attenuation coefficient; The virtual stiffness threshold is multiplied by the target exponent value to obtain the real-time virtual stiffness; where the target exponent value is the product of the damping coefficient with the natural constant as the base and the exponent of the exponent value being negative and the cabin vibration energy. The real-time virtual stiffness is used as the current stiffness parameter in the variable stiffness impedance control mode.

6. The method according to claim 2, characterized in that, The process for determining the dynamic threshold is as follows: Obtain the preset basic contact force threshold and acceleration coupling coefficient; Calculate the product of the acceleration coupling coefficient and the real-time amplitude of the acceleration data, and add this product to the basic contact force threshold to obtain the dynamic threshold.

7. A high-precision adaptive control device for a wind turbine nacelle robotic arm with a swaying base, characterized in that, The device includes: The acquisition module is used to acquire acceleration data of the wind turbine nacelle, temperature data of the robotic arm linkage, and current data of the joint motor. The calculation and processing module is used to perform frequency domain analysis and filtering separation on the acceleration data to obtain the disturbance acceleration; determine the inertial disturbance torque corresponding to each joint based on the disturbance acceleration; convert the inertial disturbance torque into the corresponding compensation current value, and superimpose the compensation current value into the driving current of the corresponding joint to obtain the compensated driving command; perform calculations based on temperature data, current data, link material parameters and reference temperature to obtain the actual length of the link, and update the kinematic parameters of the robot arm based on the actual length of the link to obtain the corrected kinematic model; The control module is used to generate drive commands for each joint based on the compensated drive commands and the corrected kinematic model, so as to control the movement of the robotic arm.

8. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes one or more computer instructions, which, when executed by a computer, perform the method as described in any one of claims 1 to 6.

Images