A solar photovoltaic panel attitude self-adaptive control system

Abstract

The present application relates to the technical fields of solar photovoltaic power generation and automatic control, in particular to a kind of posture self-adaptive control system based on solar photovoltaic panel, comprising: environmental manifold perception and modal identification step: collection light field distribution and mechanical disturbance signal, calculate light intensity spatial gradient and time variance to divide environmental modal;Virtual impedance model construction step: construct second-order mechanical impedance model, respond to light intensity gradient and adjust virtual stiffness, respond to disturbance high-frequency component and adjust virtual damping;Gravity bias and backlash compensation step: analytical gravity moment component constructs asymmetric dead zone bias, generates anti-backlash instruction using pre-tightening force generated by gravity;Collaborative execution and closed-loop iteration step: fusion control law and bias instruction drive actuator, monitor energy output and mechanical loss to update parameters;The present application solves the contradiction between tracking accuracy and mechanical reliability, effectively eliminates backlash shock.

Description

Technical Field

[0001] This invention relates to the field of solar photovoltaic power generation and automatic control technology, specifically to an adaptive control system based on the attitude of a solar photovoltaic panel. Background Technology

[0002] In the daily operation of a photovoltaic power station, the photovoltaic tracking system adjusts the attitude of the photovoltaic panels in real time through the drive mechanism to achieve precise tracking of the sun's trajectory; while the power generation gain and mechanical life of the system largely depend on the adaptability of the control strategy to environmental disturbances under complex weather conditions.

[0003] In traditional methods, photovoltaic trackers often employ high-gain rigid position servo control, treating the support as an ideal rigid body. When faced with gusts of wind or backlash in the transmission components, they cannot effectively suppress mechanical resonance, resulting in low wind disturbance stability of the system. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a solar photovoltaic panel attitude adaptive control system. Specifically, the technical solution of this invention includes:

[0005] Environmental manifold perception and modal identification unit: collects light field distribution data and disturbance signals of mechanical transmission chain, calculates light intensity spatial gradient operator and time variance statistics; based on preset manifold mapping rules, it divides the current working condition into direct light dominant mode, diffuse light dominant mode or strong wind disturbance mode, and generates environmental modal status words;

[0006] Virtual impedance model adaptive building block: Based on the environmental modal state word, a second-order mechanical impedance model containing virtual stiffness parameters and virtual damping parameters is constructed; the virtual stiffness parameters are dynamically adjusted in response to the change of the light intensity spatial gradient operator, and the virtual damping parameters are dynamically adjusted in response to the high-frequency components of the disturbance signal, generating a variable impedance control law;

[0007] Gravity bias and backlash compensation unit: Analyze the gravity torque component under the current attitude, construct the asymmetric dead zone bias threshold; apply unidirectional preload using the gravity torque component to force the transmission gear to maintain a unilateral contact state, and generate anti-backlash bias command;

[0008] Multi-source collaborative execution and closed-loop iterative unit: integrates variable impedance control law and anti-backlash bias command to calculate target action torque; drives the actuator to respond to target action torque, and monitors net energy output gain and mechanical loss index in real time to trigger closed-loop update of control parameters.

[0009] Preferably, based on preset manifold mapping rules, the method for classifying the current operating condition into a direct-light-dominated mode, a diffuse-light-dominated mode, or a strong-wind-disturbance mode includes: if the light intensity spatial gradient operator is higher than a preset gradient threshold and the temporal variance statistic is lower than a preset fluctuation threshold, then it is determined to be a direct-light-dominated mode, and a high-precision tracking strategy is locked; if the light intensity spatial gradient operator is lower than a preset gradient threshold and the temporal variance statistic is lower than a preset fluctuation threshold, then it is determined to be a diffuse-light-dominated mode, and the strategy is switched to horizontal positioning or maximum celestial projection; if the temporal variance statistic is higher than a preset fluctuation threshold, then it is determined to be a strong-wind-disturbance mode, and a compliant unloading strategy is activated; finally, the determination result is encoded to generate an environmental mode status word.

[0010] Preferably, the method of dynamically adjusting the virtual stiffness parameter in response to changes in the light intensity spatial gradient operator and dynamically adjusting the virtual damping parameter in response to high-frequency components of the disturbance signal includes: in the direct-dominant mode, establishing a positive correlation mapping between the virtual stiffness parameter and the light intensity spatial gradient operator to improve the system's response sensitivity to the target trajectory; in the strong wind disturbance mode, establishing a positive correlation mapping between the virtual damping parameter and the amplitude of the disturbance signal to reduce the position loop gain and increase the virtual viscous damping; when an oscillation characteristic of a preset frequency band is detected in the disturbance signal, it is determined to be backlash nonlinear flutter, and the virtual damping parameter is immediately increased in a stepwise manner until the oscillation characteristic decays to a safe range.

[0011] Preferably, the method of applying a unidirectional preload using the gravitational torque component to force the transmission gear to maintain a unilateral contact state includes: calculating the projection of the lever arm of the photovoltaic panel's center of gravity relative to the rotation axis in real time to obtain the passive restoring torque generated by gravitational potential energy; if the target action direction is opposite to the direction of the passive restoring torque, then superimposing a start-up compensation amount that can overcome the gravitational potential energy threshold into the target action torque; if the target action direction is the same as the direction of the passive restoring torque, then limiting the output rate of the target action torque, using gravity as a driving source to maintain continuous contact of the gear tooth surfaces, and generating an anti-backlash command.

[0012] Preferably, the method for acquiring light field distribution data and disturbance signals of mechanical transmission chains includes: acquiring differential voltage signals through a four-quadrant photosensitive array as light field distribution data; acquiring angular velocity and acceleration data through an inertial measurement unit, and simultaneously sampling the current ripple signal of the actuator; performing spectral analysis on the current ripple signal to separate the external disturbance component representing wind load impact and the internal friction component representing mechanical wear, and combining them to form the disturbance signal of the mechanical transmission chain.

[0013] Preferably, the method for real-time monitoring of net energy output gain and mechanical loss indicators to trigger closed-loop updates of control parameters includes: calculating the power generation increment within the current action cycle and deducting the energy consumption of the actuator to obtain the effective net gain; estimating the mechanical loss cost based on the action amplitude and load stress using a preset fatigue life model; constructing a cost function with the objectives of maximizing the effective net gain and minimizing the mechanical loss cost; and correcting the basic coefficients in the second-order mechanical impedance model to complete the closed-loop update of control parameters if the value of the cost function does not reach the preset optimization threshold.

[0014] Preferably, the generation of the variable impedance control law further includes: setting the virtual stiffness parameter to zero or a minimum value under the diffuse dominant mode, and setting the virtual damping parameter to a critical damping value; prohibiting the actuator from performing high-frequency fine-tuning actions, and only allowing a single large-step repositioning action when the integral value of the light intensity spatial gradient operator exceeds the preset dead zone accumulation threshold.

[0015] Preferably, the system also includes: an abnormal operating condition fuse unit: used to monitor the peak value of the disturbance signal in real time; if the peak value of the disturbance signal exceeds the preset structural damage threshold, the current variable impedance control law is forcibly bypassed; and the maximum damping lock-in command or the downwind stop command is directly output to prioritize the physical integrity of the mechanical structure.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. This invention achieves accurate quantification and classification of complex meteorological environments in photovoltaic power plants through an environmental manifold sensing and modal identification unit. The system uses a four-quadrant photosensitive array to collect light field distribution data and calculate the spatial gradient of light intensity. At the same time, it integrates the inertial measurement unit and motor current ripple signal to construct the disturbance signal of the mechanical transmission chain. By calculating the time variance statistics, it can decouple the illumination characteristics and mechanical disturbance characteristics from the multi-dimensional state space. Through the preset manifold mapping rules, the system can deterministically classify the current operating condition into direct sunlight-dominated, diffuse sunlight-dominated, or strong wind disturbance modes. In particular, through frequency domain analysis, it effectively separates the external disturbance component representing wind load impact from the internal friction component representing mechanical wear, solving the problem that traditional photovoltaic trackers cannot distinguish between strong wind and shaft friction. This provides a decision basis with clear physical meaning for the subsequent switching of control strategies and significantly reduces the blindness of operation and maintenance inspections.

[0018] 2. This invention overcomes the limitations of traditional rigid position servo systems by adaptively constructing units using a virtual impedance model. It establishes a second-order mechanical impedance model with variable parameters, which can dynamically adjust the virtual stiffness and virtual damping parameters according to the environmental modes. In the direct sunlight mode, a positive correlation mapping between virtual stiffness and light intensity gradient is established to improve tracking response sensitivity. In the strong wind mode, a positive correlation mapping between virtual damping and disturbance intensity is established to suppress jitter. In particular, for backlash nonlinear flutter, by monitoring the oscillation characteristics of a preset frequency band and stepwise increasing the virtual damping, high-frequency vibration energy in the mechanical gap can be quickly absorbed. In addition, in the diffuse dominant mode, a zero-stiffness sleep combined with an event-driven stiffness reuse strategy is adopted, which not only achieves compliant energy saving when stationary but also ensures the execution power of repositioning actions, realizing the system's combination of stiffness and flexibility under different weather conditions.

[0019] 3. This invention cleverly utilizes the gravitational torque of the photovoltaic panel itself as a natural backlash elimination preload through gravity bias and backlash compensation units, effectively solving the backlash oscillation problem commonly found in low-cost reducers. The system analyzes the gravitational torque components under the current posture in real time, constructs an asymmetric dead zone bias threshold, and superimposes the start-up compensation amount under anti-gravity conditions to prevent falls. Under forward gravity conditions, it limits the output rate and generates a reverse braking torque, forcing the transmission gears to always maintain a single-sided contact state. This asymmetric control strategy achieves zero-backlash transmission without increasing the hardware cost of dual-gear backlash elimination boxes or spring mechanisms, eliminates mechanical impact noise during large-angle tracking in the morning and evening, and significantly extends the fatigue life of mechanical transmission components.

[0020] 4. This invention constructs a dual guarantee mechanism that balances economic benefits and structural safety through multi-source collaborative execution, closed-loop iterative units, and abnormal operating condition circuit breakers. The system monitors net energy output gain and mechanical loss indicators in real time, constructs a cost function that includes the increase in power generation and the cost of mechanical wear, and automatically optimizes control parameters through a bidirectional correction strategy when mechanical loss dominates or power generation gain is insufficient, thereby maximizing the comprehensive benefits throughout the entire life cycle. At the same time, the system has the capability to circuit breakers under abnormal operating conditions. When the peak value of the disturbance signal exceeds the structural damage threshold, the system forces the bypass variable impedance control law to execute the maximum damping lockout or downwind shutdown command first, ensuring the physical integrity of the mechanical structure in extreme and uncontrollable environments such as typhoons, reflecting the design philosophy of survival first. Attached Figure Description

[0021] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0022] Figure 1 This is a structural diagram of the system of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. Example 1:

[0024] Please see Figure 1 An adaptive attitude control system based on solar photovoltaic panels, the system comprising:

[0025] The environmental manifold perception and modal identification unit is used to collect light field distribution data and disturbance signals of mechanical transmission chain, calculate light intensity spatial gradient operator and time variance statistics, and divide the current working condition into direct light dominant mode, diffuse light dominant mode or strong wind disturbance mode based on preset manifold mapping rules, and generate environmental modal status words.

[0026] The virtual impedance model adaptive building unit is used to construct a second-order mechanical impedance model containing virtual stiffness parameters and virtual damping parameters based on the environmental modal state word. It dynamically adjusts the virtual stiffness parameters in response to the change of the light intensity spatial gradient operator and dynamically adjusts the virtual damping parameters in response to the high-frequency components of the disturbance signal, thereby generating a variable impedance control law.

[0027] The gravity offset and backlash compensation unit is used to analyze the gravity torque component under the current attitude, construct the asymmetric dead zone offset threshold, and apply a unidirectional preload using the gravity torque component to force the transmission gear to maintain a unilateral contact state and generate an anti-backlash offset command.

[0028] The multi-source collaborative execution and closed-loop iteration unit is used to integrate the variable impedance control law and the anti-backlash bias command, calculate the target action torque, drive the actuator to respond to the target action torque, and monitor the net energy output gain and mechanical loss index in real time, triggering the closed-loop update of the control parameters.

[0029] In terms of physical hardware implementation, this system utilizes an industrial bus, such as... 485 or The bus connects the four-quadrant photosensitive array and the inertial measurement unit. The current sampling module integrated in the motor driver is connected to the core processor; the core processor analyzes the electrical signals from the aforementioned sensors, generates control commands according to the environmental manifold perception logic, and ultimately... Pulses or communication messages drive the actuator to move, forming a complete mechatronics control loop;

[0030] This embodiment details the specific architecture and operating mechanism of the aforementioned system. The system aims to resolve the contradiction between tracking accuracy and mechanical reliability faced by traditional photovoltaic trackers under complex meteorological conditions, such as intermittent cloud cover or strong gusts, particularly addressing the backlash oscillation problem of rigid transmission structures under random wind loads. The environmental manifold perception and modal identification unit introduces the concept of environmental manifold, which is the coupled set of the light field (light distribution characteristics) and the force field (mechanical disturbance characteristics) in a multidimensional state space. Mathematically, it is explicitly defined by the light intensity gradient operator. With time variance statistic The two-dimensional state vector formed ;

[0031] To ensure that those skilled in the art can implement this, the calculation method for the key input is specified here: the light intensity spatial gradient operator. The ratio of the absolute value of the east-west difference to the total light intensity using a four-quadrant photosensitive array, i.e. Time variance statistic The variance of the mechanical transmission chain disturbance signal within a preset sliding window is used, i.e. This unit uses multi-source sensor fusion technology to construct a real-time description of the current environmental state. Based on this, the core of the virtual impedance model adaptive construction unit lies in the introduction of a variable impedance control law. That is, the photovoltaic support is no longer considered a rigid position servo system, but rather simulated as a virtual spring-damped system with variable parameters. To achieve this control, this embodiment constructs the following second-order mechanical impedance model:

[0032] ;

[0033] in, The reference torque command output by the variable impedance control law originates from the controller calculation and has the physical meaning of driving torque, with the unit being Newton-meters.

[0034] Virtual stiffness parameter, derived from real-time mapping of light intensity spatial gradient, physically represents the system's correction force for tracking errors, and is measured in Newton-meters per radian.

[0035] The target reference angle is derived from astronomical algorithms, and its physical meaning is the theoretical position of the sun, measured in radians.

[0036] The actual measured angle comes from an angle sensor and its physical meaning is the actual posture of the photovoltaic panel, measured in radians.

[0037] Virtual damping parameter, derived from the mapping of high-frequency components of the disturbance signal, physically represents the system's ability to dissipate external shocks, and is measured in Newton-meter-seconds per radian.

[0038] Actual angular velocity, derived from gyroscope data, is physically represented as the rate of rotation, measured in radians per second.

[0039] To address the issue of sufficient disclosure regarding the adaptive construction of virtual impedance models and manifold mapping rules, this embodiment clarifies the specific algorithm implementation logic:

[0040] The specific operator for environmental manifold mapping: setting the gradient threshold For example, 0.2, compared to the variance threshold. For example, 0.01; the judgment logic is: if and Generate status word 01; if and Generate status word 00; if This generates state word 10; this logic operator implements a deterministic mapping from sensor data to modal state words.

[0041] Functional expression for virtual impedance parameter: Virtual stiffness With light intensity spatial gradient operator They follow a linear enhancement function:

[0042] ;

[0043] in, Based on the stiffness, To adjust the gain; virtual damping High-frequency components of the disturbance signal Follow the response function Therefore, the input parameters and The above specific formula is explicitly included in the second-order mechanical impedance model. The reproducibility of the control law was ensured during the calculation.

[0044] The gravity bias and backlash compensation unit aims to solve the backlash problem commonly found in low-cost reducers. Its core strategy is to use the gravitational torque of the photovoltaic panel itself as the backlash elimination preload. By analyzing the gravitational torque components, the projected torque of gravity on the rotation axis is calculated based on the current pitch angle and the position of the photovoltaic panel's center of mass. The multi-source collaborative execution and closed-loop iteration unit is responsible for the final execution and optimization. It superimposes the output of the variable impedance control law with the anti-backlash bias command to obtain the final target driving torque of the drive motor, and monitors the net energy output gain and mechanical loss indicators of the system in real time. Example 2:

[0045] Based on the preset manifold mapping rules, the current working condition is further specified into different modal steps;

[0046] Based on preset manifold mapping rules, the methods for classifying the current operating condition into direct sunlight-dominated mode, diffused sunlight-dominated mode, or strong wind disturbance mode include:

[0047] If the light intensity spatial gradient operator is higher than the preset gradient threshold and the time variance statistic is lower than the preset fluctuation threshold, it is determined to be the direct-light dominant mode, and the high-precision tracking strategy is locked.

[0048] If the light intensity spatial gradient operator is lower than the preset gradient threshold and the time variance statistic is lower than the preset fluctuation threshold, it is determined to be the diffuse dominant mode, and the strategy is switched to horizontal positioning or maximum celestial projection.

[0049] If the time variance statistic is higher than the preset fluctuation threshold, it is determined to be a strong wind disturbance mode, and the compliant unloading strategy is activated.

[0050] Finally, the judgment result is encoded to generate an environment modal status word.

[0051] This embodiment further defines the specific logic of environmental modal identification and the calculation method of statistics to solve the problem of unclear variable definitions; it defines a light intensity spatial gradient operator. With time variance statistic The specific mathematical form; The calculation is the ratio of the absolute value of the east-west light intensity difference in the four-quadrant photosensitive array to the total light intensity. The formula is as follows:

[0052] ;

[0053] in, These represent the light intensity signal values ​​collected by the photosensitive units in the four quadrant photosensitive array corresponding to the east, west, north, and south directions. To prevent overflow during division operations due to the total light intensity approaching zero at night or in extremely low light conditions, this embodiment sets an effective threshold for the total light intensity. That is, 2% of the sensor's full scale, when it detects At that time, forced placement It retains the modality determination from the previous time step and does not perform division calculations; regarding the time variance statistic... This embodiment defines it as the energy statistical characteristics of the mechanical disturbance signal, and the calculation formula is as follows:

[0054] ;

[0055] in, The normalized mechanical disturbance signal is derived from the current ripple component of Example 5. Acceleration components collected by the inertial measurement unit The weighted fusion yields, i.e. To ensure that the physical meaning of signal fusion is clear and the computation is convergent, this embodiment sets... ,satisfy The normalization constraint is set, with an emphasis on the sensitivity of the current signal, while also setting... To implement the hardware overcurrent protection threshold of the motor driver, such as 10A, For example, the upper limit of the acceleration range of the inertial measurement unit. This allows heterogeneous sensor data to be uniformly mapped to the dimensionless interval [0,1]. During the initialization phase, the system automatically reads the current hardware register values ​​of the driver and sensors, and automatically completes the process. and Parameter matching ensures normalization operator Amplitude consistency across different hardware specifications; This is the length of the sliding window, ranging from 50 to 100. This is the arithmetic mean of the normalized mechanical disturbance signal within the sliding window, i.e. This is used to eliminate the DC component in the signal, ensuring that the variance only reflects the fluctuation energy.

[0056] Based on the above definitions, this embodiment uses the following threshold logic tree for modality partitioning:

[0057] Direct-dominant mode determination: When detected That is, the preset gradient threshold ,and That is, the preset fluctuation threshold When the background noise level corresponds to the windless state, it indicates that the illumination directionality is strong and the mechanical structure is stable, and the system locks in a high-precision tracking strategy.

[0058] Diffuse dominant mode determination: when and When the light is uniform, such as in cloudy weather and a stable environment, it is determined to be the diffuse dominant mode, and the strategy is switched to horizontal positioning or maximum celestial projection.

[0059] Strong wind disturbance mode determination: Regardless of lighting conditions, once detected... This indicates that the external wind load or mechanical oscillation is severe, and it is immediately identified as a strong wind disturbance mode, forcibly activating the compliant unloading strategy;

[0060] The above determination result is encoded into a 2-bit binary status word, where 00 represents diffuse radiation, 01 represents direct radiation, and 10 represents strong wind, and is transmitted to the downstream control unit. This embodiment corrects the problem of mismatch between acceleration sign and unit in the original embodiment by introducing explicit weight definition and disturbance variance calculation with normalized physical range, eliminates the ambiguity of parameter definition, and ensures the portability of mode switching logic under different sensor hardware specifications. Example 3:

[0061] The methods for dynamically adjusting virtual stiffness parameters in response to changes in the spatial gradient operator of light intensity, and for dynamically adjusting virtual damping parameters in response to high-frequency components of disturbance signals, include:

[0062] Under the direct-light-dominated mode, a positive correlation mapping between virtual stiffness parameters and light intensity spatial gradient operators is established to improve the system's response sensitivity to target trajectories.

[0063] Under strong wind disturbance mode, a positive correlation mapping between virtual damping parameters and disturbance signal amplitude is established to reduce position loop gain and increase virtual viscous damping.

[0064] When an oscillation characteristic of a preset frequency band is detected in the disturbance signal, it is determined to be backlash nonlinear flutter. The virtual damping parameter is immediately increased in a stepwise manner until the oscillation characteristic decays to a safe range.

[0065] This embodiment elaborates on the virtual stiffness parameters. With virtual damping parameters The dynamic adjustment mechanism is the core algorithm for achieving adaptive control. In stiffness adjustment under the direct sunlight dominant mode, the system establishes the following positive correlation mapping to quickly respond to changes in the sun's position:

[0066] ;

[0067] in, The basic stiffness coefficient, whose physical meaning is the static stiffness of the system, is measured in Newton-meters per radian. Its typical value ranges from 5,000 to 20,000. ; This is the stiffness gain factor, which physically represents the adjustment sensitivity. It is a dimensionless constant with typical values ​​ranging from 0.5 to 2.0. The spatial gradient operator for real-time light intensity acquisition, physically representing the directional intensity of illumination, is a dimensionless ratio. In damping adjustment under strong wind disturbance modes, to suppress wind-induced shaking, the system establishes the following mapping:

[0068] ;

[0069] in, The basic damping coefficient, physically representing the inherent damping of the system, is measured in Newton-meter-seconds per radian. Its typical value ranges from 100 to 500. ; is the damping sensitivity factor, which physically represents the response gain to wind load. It is a dimensionless constant with a value range of 3.0 to 5.0. The normalized perturbation intensity exponent is a dimensionless parameter; the gain is adjusted. With damping sensitive factor The system supports online calibration based on actual power plant measurement data, and after completing a full closed-loop iteration cycle, the optimized base coefficients will be applied. and Write to non-volatile memory to enable continuous evolution of control strategies;

[0070] To address the calculation difficulties caused by the inconsistency between the current signal (unit: amperes) and the acceleration signal (unit: meters per second squared) in Example 5, this example employs the following normalization formula:

[0071] ;

[0072] in, This indicates the calculation of the root mean square value within the sliding window. This represents the fluctuation value after filtering out the DC component from the current ripple. For high-frequency vibration components collected by the accelerometer; and These are the upper limits of the corresponding sensor ranges; and These are the weighting coefficients for the current ripple component and the acceleration component, respectively. To balance the contributions of the two types of sensor data and maintain consistency in the system's internal parameter settings, this embodiment sets... This weight allocation is consistent with the weight setting of the normalized disturbance signal in Example 2, both focusing on the sensitivity of the current signal and satisfying the following conditions. The constraints; this processing ensures The damping parameter is always in the [0,1] interval, making the damping parameter The regulation has clear physical boundaries;

[0073] To suppress backlash nonlinear chatter, when the system detects a preset frequency band (typically 0.5) in the disturbance signal through spectral analysis... Up to 5 When the oscillation characteristics of the mechanical resonance zone are detected, it is determined to be backlash flutter. It should be noted that the preset frequency band is a reference range set based on the resonance frequency of a typical mechanical structure. In actual engineering applications, the preset frequency band can be parameterized and calibrated according to the inherent frequency characteristics of a specific type of photovoltaic bracket through offline impact test or online frequency sweep test to adapt to mechanical transmission chains of different materials and geometric dimensions.

[0074] The specific adjustment process is as follows: In each control cycle... , here Defined as the discrete sampling and execution time step of the digital controller, with a value of 10. Up to 50 Within this framework, based on the latest sliding sampling window data, the energy spectral density of the oscillation frequency band is calculated. To ensure the physical meaning of the frequency domain analysis, the sampling window length... Sampling frequency With frequency resolution satisfy In this embodiment, when and At that time, the frequency resolution is approximately It can cover 0.5 Up to 5 The need for resonance analysis The calculation formula is:

[0075] ;

[0076] in, Let be the amplitude of the Fast Fourier Transform of the disturbance signal; if Exceeding the safety threshold If the threshold is set to 5% of the energy spectral density corresponding to the system's rated torque, then execution will proceed.

[0077] ;

[0078] in, The preset damping step value, such as critical damping. 5%, here defined as critical damping. The calculation formula is:

[0079] ;

[0080] in, The equivalent rotational inertia of the system is expressed in units of 1. ; The current virtual stiffness parameter, in units of Critical damping With virtual stiffness parameters The adaptive adjustment is recalculated in real time to ensure that the system can maintain the preset damping ratio characteristics under different stiffness conditions; dimensional analysis shows that... Equivalent to The calculation results are consistent with the virtual damping parameters. unit Complete consistency ensures the uniformity of the dimensions of physical quantities in the control law and the accuracy of calculations; this dimensional alignment based on physical reality eliminates floating-point operation deviations that may occur when the software algorithm is ported to different controller platforms, ensuring that the virtual damping parameter D is consistent across Newton-meter-seconds per radian. Numerical stability under dimensions;

[0081] To prevent the system from being in an overdamped state for an extended period, this embodiment also introduces a damping release mechanism: when a damping release mechanism is detected... continuous One cycle, for example, 200 cycles, which is approximately 2-10 seconds below When damping attenuation logic is executed, its calculation formula is as follows:

[0082] ;

[0083] in, The theoretical damping value is calculated based on the current wind load mapping. The preset linear decay step size is set to the damping step value. This allows for a 50% reduction in the control parameters, thus enabling flexible reset of the control parameters. Example 4:

[0084] Further details of the step of applying a unidirectional preload using the gravitational torque component to force the transmission gear to maintain a unilateral engagement state;

[0085] The method includes: calculating the projection of the lever arm of the photovoltaic panel's center of gravity relative to the rotation axis in real time to obtain the passive restoring torque generated by gravitational potential energy;

[0086] In response to the fact that the direction of the target action is opposite to the direction of the passive restoring torque, an initiation compensation amount that can overcome the gravitational potential energy threshold is superimposed on the target action torque;

[0087] In response to the fact that the direction of the target motion is the same as the direction of the passive restoring torque, the output rate of the target motion torque is limited, gravity is used as the driving source to maintain continuous contact of the gear tooth surface and generate an anti-backlash command.

[0088] This embodiment details the specific implementation method of applying unidirectional preload using the gravitational torque component; the system calculates the passive restoring torque generated by gravity in real time. The calculation formula is as follows:

[0089] ;

[0090] in, The total mass of the photovoltaic module is derived from a preset constant, and its physical meaning is the load mass, with the unit being kilograms.

[0091] Gravitational acceleration, derived from the geographic location constant, is physically defined as the intensity of the gravitational field, and is measured in meters per second squared.

[0092] The length of the lever arm from the center of mass to the axis of rotation is derived from mechanical structural parameters. Its physical meaning is the torque arm length, and the unit is meters.

[0093] The current tilt angle is obtained from real-time sensor data and its physical meaning is the attitude of the plate surface, measured in radians.

[0094] The system executes the backlash-resistant instruction generation logic; under reverse gravity conditions, i.e., the target's direction of motion is... Conversely, the system adds a starting compensation amount to the target torque. In order to clarify The specific values ​​are set in this embodiment to ensure reliable startup.

[0095] ;

[0096] in, The static friction torque of the system is obtained through offline identification. The specific identification steps are as follows: In a windless environment, control the actuator motor at an extremely low speed, such as less than... Rotate at a constant speed, record the average value of the motor output torque at this time, and subtract the gravitational torque component under the current posture. The resulting residual is the static friction torque of the system. This ensures that the motor output is sufficient to overcome the sum of gravitational potential energy and static friction, preventing a drop at startup; under gravity-driven conditions, i.e., the target movement direction is... At the same time, since gravity itself tends to pull the photovoltaic panel to accelerate, it can easily cause the gear to disengage from the contact surface, that is, the tooth surface jumps from the active side to the passive side and causes an impact.

[0097] Therefore, this embodiment limits the motor's output rate, and even outputs a reverse braking torque, so that the photovoltaic panel's movement speed is slightly lower than its natural falling speed due to gravity; specifically, it generates a negative anti-backlash bias command and sets the bias coefficient. The value ranges from 1.05 to 1.1, making the total holding torque such that... This forces the bearing surface of the gear to always remain on the same side, utilizing gravity as a natural backlash-eliminating spring;

[0098] This embodiment employs an asymmetric control strategy, achieving zero-backlash transmission through algorithms alone without using expensive dual-gear backlash elimination boxes or spring preload mechanisms, thus significantly reducing hardware costs. In scenarios involving large-angle tracking in the early morning and late evening, it effectively eliminates mechanical impact noise and improves the system's operational stability. Example 5:

[0099] Further details on the steps for acquiring optical field distribution data and disturbance signals from the mechanical transmission chain; acquisition methods include:

[0100] Differential voltage signals are acquired using a four-quadrant photosensitive array and used as optical field distribution data.

[0101] Angular velocity and acceleration data are acquired through an inertial measurement unit, and the current ripple signal of the actuator is sampled simultaneously.

[0102] Spectral analysis of the current ripple signal separates the external disturbance component representing wind load impact and the internal friction component representing mechanical wear, which are then combined to form the disturbance signal of the mechanical transmission chain.

[0103] This embodiment specifically illustrates the acquisition and processing methods for optical field distribution data and mechanical transmission chain disturbance signals. For optical field data acquisition, the system uses a four-quadrant photosensitive array, comparing the photocurrent differences between diagonal quadrants to output a differential voltage signal. This signal is then low-pass filtered, for example, with a cutoff frequency set to 2 kHz. After processing, it serves as the light field distribution data; in terms of mechanical disturbance signal acquisition and separation, the system synchronously samples the angular velocity of the inertial measurement unit. acceleration and the armature current of the motor In order to effectively distinguish between high-frequency mechanical wear signals and low-frequency wind load signals, this embodiment sets the current sampling frequency. At least 1000 This is to satisfy the Nyquist sampling theorem's requirement for capturing high-frequency features;

[0104] Due to motor current The information includes both external wind load factors and internal friction factors. In this embodiment, the current ripple signal is subjected to fast Fourier transform spectrum analysis using a sliding time window of 1024 points, which is consistent with the window length used for frequency domain analysis in the embodiment 3, in order to avoid parameter definition conflicts.

[0105] During this process, the system extracts the low-frequency band and sets the cutoff frequency to 0.8. This frequency band exhibits random amplitude characteristics and is used as a component of wind load impact; simultaneously, the mid-to-high frequency band is extracted, with a frequency range set at 0.8. Up to 200 This frequency band includes not only 0.5 of Example 3. Up to 5 The backlash characteristics also include those greater than 10 times the frequency multiple of the motor speed. The mechanical wear component; the separated wind load component is combined with the data from the inertial measurement unit to form the disturbance signal of the mechanical transmission chain; in order to clarify the input signal source of different control loops and conform to rule 1, this embodiment sets the signal output interface to dual-channel mode:

[0106] aisle Output full-band weighted perturbation value For use in the modal variance calculation of Example 2; Channel Output high-frequency components after high-pass filtering The cutoff frequency of the high-pass filter is greater than 0.8. To address the issue of undisclosed high-pass filter type and parameters preventing programmable implementation, this embodiment explicitly specifies that the filter employs a second-order Butterworth high-pass topology with a sampling frequency of... Cutoff frequency Its digital implementation follows a difference equation, and the calculation formula is as follows:

[0107] ;

[0108] The coefficients are calculated based on the bilinear transformation method: Let ,but ;

[0109] This algorithm ensures the unique determinism of extracting high-frequency components from the original signal, which is then used for calculation by the virtual impedance model construction unit in Example 3. The index is used to avoid interference from low-frequency wind load signals in the determination of backlash flutter.

[0110] This embodiment achieves decoupling of disturbance sources through frequency domain analysis. Traditional current detection cannot distinguish between strong winds and shaft friction, while this solution enables the control system to increase damping for wind loads and may trigger maintenance alarms for mechanical wear, i.e., increased friction components. This refined diagnosis can significantly reduce the blindness of operation and maintenance inspections in remote unattended power plants. Example 6:

[0111] Real-time monitoring of net energy output gain and mechanical loss indicators, triggering closed-loop updates of control parameters, includes the following methods:

[0112] Calculate the power generation increment within the current action cycle and deduct the energy consumption of the actuator to obtain the effective net gain; estimate the mechanical loss cost based on the action amplitude and load stress using a preset fatigue life model.

[0113] Construct a cost function that aims to maximize effective net gain and minimize mechanical loss cost;

[0114] If the value of the cost function does not reach the preset optimization threshold, the basic coefficients in the second-order mechanical impedance model are corrected to complete the closed-loop update of the control parameters.

[0115] This embodiment details a bidirectional parameter update mechanism based on a cost function in a closed-loop iterative unit, aiming to address logic loopholes caused by unidirectional adjustment; the system calculates the effective net gain. With mechanical wear and tear costs To ensure the reproducibility of the calculation process, this embodiment clarifies that... Sources of data collection for each physical quantity: The real-time power generation of the photovoltaic string is determined by... String voltage sampled by the controller With string current The product is obtained, that is ; The real-time operating energy consumption of the actuator is determined by the motor driver bus voltage. With the sampled phase current The product is obtained, that is The integration process in the controller is implemented using discrete time steps. Implementation of cumulative summation:

[0116] ;

[0117] ;

[0118] in, For load stress, this embodiment eliminates the black-box description calculated from the current and establishes a clear physical mapping formula:

[0119] ;

[0120] in, The torque constant of the motor is expressed in units of 1000 ppm. , This is the overall transmission ratio of the reducer. This is the root mean square value of the current during the operating cycle. The effective bending section modulus of the final stage transmission gear, in units of ; The fatigue index is set to a constant of 3.0 for steel. The mechanical wear coefficient, characterizing the unit amplitude of motion, is determined by the life-cycle cost model of the transmission mechanism, and the calculation formula is as follows:

[0121] ;

[0122] in, The cost of replacing the speed reducer, for example, is 1500 yuan. The total operating life in radians under rated load, for example radians, therefore in this embodiment Values Each radian ensures that in the cost function The item has a clearly defined monetary unit, with the unit being yuan; Given the yield strength stress of the material in the mechanical transmission components; construct a cost function with the objectives of maximizing effective net gain and minimizing mechanical loss cost:

[0123] ;

[0124] in, The economic conversion factor per unit of power generation gain, such as the feed-in tariff, in yuan. Used to convert energy gain into mechanical loss cost. Same monetary units; The weighting for operational risk penalties is defined by the following formula:

[0125] ;

[0126] in, This represents the average amount of power generation loss caused by a single unplanned outage. The total labor cost for repairs, This represents the unit price for core component procurement; this formula makes implicit maintenance risks explicit. If the cost function... The value did not reach the preset optimization threshold. This triggers the correction of the fundamental coefficients in the second-order mechanical impedance model;

[0127] It should be noted here that, although In The coefficient has mapped mechanical wear to the theoretical replacement cost in yuan. However, considering the extremely high indirect losses caused by unplanned maintenance of power plants in remote desert areas, such as power generation loss due to shutdown and high labor travel expenses and special vehicle entry fees, these losses are often dozens of times the material cost of the components themselves.

[0128] Therefore, this embodiment introduces... Here, a penalty weight of 50 is used to significantly amplify the impact of mechanical loss in the cost function, forcing the optimization algorithm to prioritize low-wear strategies when not necessary, rather than just pursuing small power generation gains, thereby maximizing the overall benefits throughout the entire life cycle.

[0129] To achieve closed-loop updates of control parameters, this embodiment sets up a two-way correction strategy based on the profit-loss ratio: the system periodically calculates... Value, if Below the preset optimization value Here, The calculation is based on: 85% of the average optimal cost function value of the power station in the same period of the previous year, or based on the theoretical maximum power generation. The calculated baseline value; where, Defined as the nominal power of a photovoltaic module under standard test conditions. via real-time battery temperature and irradiance The corrected theoretical output, i.e. This ensures the objectivity and dynamic adaptability of the threshold setting, and then calculates the loss-to-benefit ratio for the current period. And execute the following branches:

[0130] Mechanical wear-prone type This indicates that the current tracking strategy is too rigid, causing wear and tear costs to exceed power generation revenue. In this case, a predetermined priority adjustment logic is executed to resolve the pseudocode ambiguity in rule 5: the current foundation stiffness coefficient is detected. Is it greater than the preset safety lower limit? The judgment condition is If the conditions are met, the foundation stiffness coefficient should be reduced first. , Mechanical stress is directly reduced by releasing the elastic dead zone; if the condition is not met, i.e., the stiffness has been reduced to the physical limit, further reduction may lead to excessive sagging due to self-weight, then maintain... Instead of keeping it unchanged, the foundation damping coefficient was increased. , Viscous damping is used to dissipate high-frequency oscillation energy;

[0131] Power generation gain deficiency type This indicates that mechanical losses are still within a controllable range, but power generation gain... Too low a value usually indicates alignment deviation due to tracking lag or insufficient stiffness; in this case, increase the foundation stiffness coefficient. To improve the system's response speed and tracking accuracy;

[0132] in, To ensure both parameter convergence speed and system stability, the adaptive iteration step size coefficient is set to a fixed constant between 0.02 and 0.05 in this embodiment, preventing excessive single update amplitude from causing control system oscillations.

[0133] Furthermore, to address the potential parameter divergence risk during the closed-loop update process, namely, infinite iteration violating Rule 5, this embodiment further defines safe saturation boundaries for the basic stiffness coefficient and damping coefficient;

[0134] ;

[0135] in, This refers to the stall torque of the motor. To ensure the angle sensor resolution and prevent excessive stiffness from causing high-frequency oscillations in the system due to quantization noise; definition:

[0136] ;

[0137] in, For the maximum gravitational torque, To prevent excessive sag due to self-weight caused by insufficient stiffness, the maximum allowable static deviation is set, such as 1 degree, and updates are performed using... The limiting logic ensures that the control parameters are always within the physically realizable stable region. Example 7:

[0138] The generation of variable impedance control laws also includes:

[0139] Under the diffuse dominant mode, the virtual stiffness parameter is set to zero or a minimum value, and the virtual damping parameter is set to the critical damping value.

[0140] The actuator is prohibited from performing high-frequency fine-tuning actions. Only when the integral value of the light intensity spatial gradient operator exceeds the preset dead zone accumulation threshold, a single large-step repositioning action is allowed.

[0141] This embodiment further refines the control strategy under the diffuse dominant mode; when the diffuse dominant mode is determined, such as on a cloudy day, the variable impedance control law executes a specific configuration; the system uses virtual stiffness parameters Setting it to zero or a minimum value, such as 1% of the base stiffness, means that the system is extremely insensitive to light intensity gradients, while simultaneously setting the virtual damping parameter... Set as the critical damping value, here is the critical damping value Defined as based on the system's fundamental stiffness and system rotational inertia The calculated theoretical value, i.e. This means that even at low stiffness, the system is in a highly damped overdamped state, making it extremely difficult for oscillations to occur.

[0142] Based on this, to prevent frequent start-stops caused by minor fluctuations in illumination, this embodiment introduces a dead-zone integral triggering mechanism to prohibit the actuator from performing high-frequency fine-tuning; only when the light intensity spatial gradient operator... The integral value exceeds the preset dead zone accumulation threshold. At that time, here Defined as minimum effective gradient With integration time window The product of The minimum reliable threshold for the rate of change of light intensity. The duration is calculated cumulatively, in seconds; for example... This ensures that only continuous changes in brightness can trigger the function, thus satisfying the requirement. The system only performs a single large-step relocation operation. To further improve the system's robustness to weather changes, the dead zone accumulation threshold is... Seasonal adjustments can be made based on historical meteorological characteristics of the current geographical location, such as increasing [the amount of something] during the cloudy summer months. To reduce the number of ineffective movements, and to appropriately reduce them in the clear, cloudless winter. To maintain fine tracking; in this process, in order to solve This leads to a logical loop problem where the system synchronously activates a stiffness pulse the instant the action command is issued, i.e., forcibly applying stiffness pulses within the action window. Upgraded to The stiffness is reset immediately after the action is completed; this is the event-driven stiffness reuse strategy detailed below.

[0143] It is important to note the control logic paradox that exists between setting the virtual stiffness parameter to zero and performing large-step relocation, namely... Time control law Unable to generate driving torque, violating rule 4, this embodiment adopts an event-driven stiffness reuse strategy; when the integral value exceeds When a relocation action is triggered, the system enters a transient forced displacement state, which lasts for a period of time. Inside( The controller forces the virtual stiffness parameters to be... Temporarily restore the basic stiffness to the direct mode ,use The generated torque drives the photovoltaic panel to rotate rapidly to a new reference position;

[0144] Once the action is completed, the position deviation is monitored in real time. When the judgment condition is met This is clearly stated. The value is 10 times the system encoder resolution, for example. and angular velocity Immediately Reset to zero or minimum value to restore diffuse sleep state; to meet the explicit control flow logic requirement of Rule 5, the specific state machine transition logic of this event-driven stiffness reuse strategy is defined as follows:

[0145] _ ( ): Continuously accumulate gradient integrals ;like Then trigger _ Jump to _ ;

[0146] _ ( _ ): Execute position closed-loop control; if Then trigger _ Jump to _ This logic ensures deterministic execution of driving actions in diffuse mode; this hybrid automaton logic ensures that the system can be as soft as cotton to conserve energy when stationary, and as rigid as steel to execute instructions when in motion. Example 8:

[0147] The system also includes:

[0148] Abnormal operating condition fuse unit, used to monitor the peak value of disturbance signal in real time;

[0149] In response to a disturbance signal peak exceeding a preset structural damage threshold, the current variable impedance control law is forcibly bypassed, and a maximum damping lockout command or a downwind shutdown command is directly output to prioritize the physical integrity of the mechanical structure.

[0150] This embodiment describes the implementation of an abnormal operating condition fuse unit, which serves as a safety barrier for the system, and the system monitors the peak value of the disturbance signal in real time. The signal primarily originates from wind load impact; simultaneously, the system has a preset structural damage threshold. This value is set based on the yield strength of the photovoltaic support, typically taken as 80% of the yield strength; the system implements a fuse bypass mechanism; in response to If the system determines that it is currently in an extremely dangerous operating condition, such as a typhoon passing through, the system will forcibly bypass the current variable impedance control law, that is, no longer calculate the virtual stiffness and damping parameters; the system will directly output the highest priority downwind shutdown command, that is, adjust the photovoltaic panel to an angle parallel to the wind direction, or the maximum damping lock command, that is, lock the motor brake and activate the electromagnetic damping brake, prioritize the physical integrity of the mechanical structure, and give up the power generation revenue.

[0151] This embodiment ensures that in extremely uncontrollable environments, the control algorithm will not cause motor overload or structural collapse due to attempts to maintain attitude; when encountering extreme weather disasters that exceed design standards, this unit embodies the design philosophy of prioritizing survival and preserves power plant assets to the greatest extent possible.

[0152] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A solar photovoltaic panel-based attitude adaptive control system, characterized in that, include: Environmental manifold sensing and modal identification unit: collects light field distribution data and disturbance signals of mechanical transmission chain, and calculates light intensity spatial gradient operator and time variance statistics; Based on the preset manifold mapping rules, the judgment result is encoded into a 2-bit binary state word to generate an environmental mode state word, where 00 represents the diffuse dominant mode, 01 represents the direct dominant mode, and 10 represents the strong wind disturbance mode. Virtual impedance model adaptive building block: Based on the environmental modal state word, a second-order mechanical impedance model containing virtual stiffness parameters and virtual damping parameters is constructed; the virtual stiffness parameters are dynamically adjusted in response to the change of the light intensity spatial gradient operator, and the virtual damping parameters are dynamically adjusted in response to the high-frequency components of the disturbance signal, generating a variable impedance control law; Gravity bias and backlash compensation unit: Analyze the gravity torque components under the current attitude and construct an asymmetric dead zone bias threshold; By applying a unidirectional preload using the gravitational torque component, the transmission gear is forced to maintain a unilateral engagement state, generating an anti-backlash command. Multi-source collaborative execution and closed-loop iterative unit: integrates variable impedance control law and anti-backlash bias command to calculate target action torque; The actuator is driven to respond to the target action torque, and the net energy output gain and mechanical loss indicators are monitored in real time, triggering closed-loop updates of control parameters; Based on preset manifold mapping rules, the current operating condition is classified into direct sunlight-dominated mode, diffuse sunlight-dominated mode, or strong wind disturbance mode in the following ways: if the light intensity spatial gradient operator is higher than a preset gradient threshold and the temporal variance statistic is lower than a preset fluctuation threshold, it is determined to be direct sunlight-dominated mode, and a high-precision tracking strategy is locked; if the light intensity spatial gradient operator is lower than a preset gradient threshold and the temporal variance statistic is lower than a preset fluctuation threshold, it is determined to be diffuse sunlight-dominated mode, and the horizontal positioning or maximum celestial projection strategy is switched; if the temporal variance statistic is higher than a preset fluctuation threshold, it is determined to be strong wind disturbance mode, and a compliant unloading strategy is activated; finally, the determination result is encoded to generate an environmental mode status word. The generation of variable impedance control law also includes: setting the virtual stiffness parameter to zero or a minimum value under the diffuse dominant mode, and setting the virtual damping parameter to a critical damping value; prohibiting the actuator from performing high-frequency fine-tuning actions, and only allowing a single large-step repositioning action when the integral value of the light intensity spatial gradient operator exceeds the preset dead zone accumulation threshold.

2. The solar photovoltaic panel attitude adaptive control system according to claim 1, characterized in that, The methods for dynamically adjusting the virtual stiffness parameter in response to changes in the spatial gradient operator of light intensity and dynamically adjusting the virtual damping parameter in response to high-frequency components of the disturbance signal include: establishing a positive correlation mapping between the virtual stiffness parameter and the spatial gradient operator of light intensity under the direct-light dominant mode to improve the system's response sensitivity to the target trajectory; establishing a positive correlation mapping between the virtual damping parameter and the amplitude of the disturbance signal under the strong wind disturbance mode to reduce the position loop gain and increase the virtual viscous damping; when oscillation characteristics of a preset frequency band are detected in the disturbance signal, it is determined to be backlash nonlinear flutter, and the virtual damping parameter is immediately increased in a stepwise manner until the oscillation characteristics decay to a safe range.

3. The solar photovoltaic panel attitude adaptive control system according to claim 1, characterized in that, The method of applying a unidirectional preload using the gravitational torque component to force the transmission gear to maintain a unilateral contact state includes: calculating the projection of the lever arm of the photovoltaic panel's center of gravity relative to the rotation axis in real time to obtain the passive restoring torque generated by gravitational potential energy; if the target action direction is opposite to the direction of the passive restoring torque, then superimposing a starting compensation amount that can overcome the gravitational potential energy threshold into the target action torque; if the target action direction is the same as the direction of the passive restoring torque, then limiting the output rate of the target action torque, using gravity as a driving source to maintain continuous contact of the gear tooth surfaces, and generating an anti-backlash command.

4. The solar photovoltaic panel attitude adaptive control system according to claim 1, characterized in that, The methods for acquiring optical field distribution data and disturbance signals of mechanical transmission chains include: acquiring differential voltage signals through a four-quadrant photosensitive array as optical field distribution data; acquiring angular velocity and acceleration data through an inertial measurement unit and simultaneously sampling the current ripple signal of the actuator; performing spectral analysis on the current ripple signal to separate the external disturbance component representing wind load impact and the internal friction component representing mechanical wear, and combining them to form the disturbance signal of the mechanical transmission chain.

5. The solar photovoltaic panel attitude adaptive control system according to claim 2, characterized in that, Real-time monitoring of net energy output gain and mechanical loss indicators triggers closed-loop updates of control parameters in the following ways: calculating the power generation increment within the current operating cycle and deducting the energy consumption of the actuator to obtain the effective net gain; estimating mechanical loss cost based on the operating amplitude and load stress using a preset fatigue life model; constructing a cost function with the objectives of maximizing effective net gain and minimizing mechanical loss cost; if the value of the cost function does not reach the preset optimization threshold, correcting the basic coefficients in the second-order mechanical impedance model to complete the closed-loop update of control parameters.

6. The solar photovoltaic panel attitude adaptive control system according to claim 1, characterized in that, The system also includes: an abnormal operating condition fuse unit: used to monitor the peak value of the disturbance signal in real time; if the peak value of the disturbance signal exceeds the preset structural damage threshold, the current variable impedance control law is forcibly bypassed; and the maximum damping lock-in command or the downwind stop command is directly output to prioritize the physical integrity of the mechanical structure.

Images