Adaptive control method of tower crane luffing-slewing linkage in high wind load environment

Abstract

The present invention relates to the field of automatic control technology, and further to a tower crane luffing-rotation linkage adaptive control method for a high wind load environment. The method comprises the following steps: Step 1: obtaining the measured average wind speed at the height of the tower crane to determine the wind-induced coupling torque acting on the tower crane's slewing shaft; Step 2: real-time calculation of the real-time equivalent moment of inertia generated by the boom deadweight, the hoisted mass and their relative positions; Step 3: obtaining the original luffing torque instruction for the tower crane's luffing motor output; Step 4: using the slewing shaft coupling moment of inertia to update the tower crane's slewing angular velocity setting value in the next control cycle in real time, thereby completing the tower crane's luffing-rotation linkage adaptive control. The present invention improves the safety, accuracy and continuous operation capability of tower cranes under complex construction conditions.

Description

Technical Field

[0001] The invention belongs to the technical field of automatic control, and in particular relates to a tower crane luffing-slewing linkage adaptive control method for a high wind load environment. Background Art

[0002] During the operation of tower cranes, wind load disturbances are always a key factor affecting their structural stability and control accuracy. Especially in high wind load environments, complex wind field characteristics such as sudden changes in wind speed, gust pulsation, and air flow shear will have a strong impact on the tower crane's boom system and hoisting system, resulting in a series of safety hazards such as increased hoist swing, delayed attitude response, and even structural overload. With the increasing demand for high-risk, high-wind area projects such as super-high-rise buildings, offshore wind power, and ultra-high voltage construction, the problem of tower cranes' adaptability to wind loads in harsh environments has become increasingly prominent, and there is an urgent need to upgrade and optimize traditional tower crane control systems from the control strategy level. Most existing tower crane control systems adopt a modular design, that is, the luffing mechanism and the slewing mechanism are independently controlled according to their own dynamic models. The typical control strategy is mainly based on PID regulation, supplemented by mechanisms such as speed feedforward or current compensation. Under normal working conditions, basic trajectory tracking and operational response can be achieved. However, when wind loads act significantly, this type of control system generally faces three problems:

[0003] First, they lack the ability to identify wind-induced disturbances in real time. Traditional control systems typically model external disturbances as small, constant-mean perturbations or white noise, ignoring the nonstationary nature of real wind fields. Especially in high-load environments where gusts and vortex structures are frequent, wind speed variations are not only sudden but also exhibit strong spatial correlation and temporal asynchrony. Single-point wind measurement devices are unable to provide effective wind speed distribution information across the entire boom system. Because the control system lacks access to complete data reflecting the spatial characteristics of the wind field, there is a response lag between control commands and actual disturbances, which can easily lead to amplified control errors and even unstable torque commands. Second, existing systems generally use fixed structural parameters for dynamic modeling, failing to dynamically adapt to changes in structural loads. During tower crane operation, the load mass, boom length, and luffing angle are constantly changing, directly impacting the overall moment of inertia and coupled torque distribution. In traditional control models, inertia parameters are typically determined during factory calibration and are not updated during operation. This static modeling method cannot reflect the dynamic structural response during actual operation, especially when the load changes greatly or the wind disturbance is severe. The original model will have large errors, resulting in reduced control accuracy and even abnormal mechanism response. Summary of the Invention

[0004] The main purpose of the present invention is to provide a tower crane luffing-slewing linkage adaptive control method for high wind load environments, thereby improving the safety, accuracy and continuous operation capability of the tower crane under complex construction conditions.

[0005] In order to solve the above problems, the technical solution of the present invention is achieved as follows:

[0006] A tower crane luffing-slewing linkage adaptive control method for high wind load environments, the method comprising:

[0007] Step 1: Obtain the measured average wind speed at the tower crane's height. Combined with the crane's slewing angular velocity, boom length, luffing angle, and slewing angle, calculate the boom's windward projected area and relative wind speed to determine the wind-induced coupling torque acting on the crane's slewing axis.

[0008] Step 2: Calculate the vertical projection distance from the tower crane's slewing center to the load based on the crane's inherent moment of inertia, boom deadweight, load mass, boom length, luffing angle, and slewing angle. Calculate the real-time equivalent moment of inertia generated by the boom deadweight, load mass, and their relative positions.

[0009] Step 3: Obtain the original luffing torque command for the tower crane luffing motor output by multiplying the real-time equivalent moment of inertia and the luffing angular acceleration command, the viscous damping effect of the luffing mechanism, the gravitational torque generated by the load, and the projection of the wind-induced coupling torque in the luffing direction;

[0010] Step 4: Using the coupled rotational inertia of the rotating shaft, the projection of the calculated wind-induced coupling torque in the rotation direction, the overturning component of the gravity generated by the hoisted mass on the rotating shaft, and the kinetic energy coupling effect caused by the linkage between the luffing angular velocity and the rotation angular velocity are converted into a correction value for the rotation angular velocity. This is used to update the rotation angular velocity set value of the tower crane in the next control cycle in real time, thereby completing the linked adaptive control of the tower crane's luffing and rotation.

[0011] Furthermore, in step 1, ultrasonic wind measuring devices are respectively arranged near the tip of the tower crane boom, the tower cap position, and at predetermined heights of multiple layers above the tower base to synchronously collect the incoming wind speed, wind direction, and turbulent pulsation intensity as raw meteorological data; the raw meteorological data is timestamped according to a unified clock and sent to the central processing unit to form a wind field raw data set covering all measuring points within each control cycle.

[0012] Furthermore, the central processing unit performs the following processes: multi-level filtering is performed on the original data set of the wind farm, specifically including: using bandpass filtering to suppress low-frequency drift and high-frequency noise, using a sliding time window to smooth gust pulsation, and removing instantaneous distortion samples according to a preset abnormality detection threshold to obtain effective wind speed and effective wind direction with stable statistical characteristics; according to the real-time amplitude angle, rotation angle and boom length of the tower crane, the coordinate transformation algorithm is called to decompose the effective wind speed into a tangential component consistent with the longitudinal direction of the boom and a normal component perpendicular to the boom plane; according to the windward projection area data of the boom The equivalent wind-exposed area under different postures is matched by table lookup, and the tangential component and the normal component are combined with the equivalent wind-exposed area to construct a comprehensive wind load field reflecting the instantaneous aerodynamic state of the tower crane; the comprehensive wind load field is mapped to the slewing axis coordinate system, and the wind-induced coupling torque acting on the slewing axis is obtained by the integration method according to the length of the effective arm from the boom to the slewing center and the position of the center of mass of the boom; the slewing angular velocity and the tangential linear velocity of the boom tip are read synchronously, and the wind-induced coupling torque is coupled and corrected in combination with the real-time wind direction deviation angle to reflect the real-time influence of the slewing motion on the relative wind speed.

[0013] Furthermore, step 2 specifically includes: reading the inherent moment of inertia of the tower body and slewing platform, the deadweight of the boom, the length of the boom, and the geometric position of the center of mass of the boom and the center of rotation as a set of static design parameters, and locking the static design parameter set once based on the factory calibration report; obtaining the real-time value of the load mass and the instantaneous position of the load in the longitudinal direction of the boom, combining the amplitude angle and rotation angle information shared in step 1, and performing coordinate transformation on the vertical projection distance of the load in the rotation plane and the action arm from the boom to the center of rotation to construct a dynamic position model of the load mass in the rotation axis coordinate system; according to the dynamic equivalence principle of the tower crane, the inherent moment of inertia of the tower body and slewing platform, the deadweight of the boom and its inertia component caused by the change of the amplitude angle, and the inertia contribution of the load mass to the rotation axis at the real-time position are accumulated separately, and the coupling inertia cross term between the boom and the center of mass of the load relative to the rotation center is introduced in the accumulation process to obtain the real-time equivalent moment of inertia.

[0014] Furthermore, the real-time value of the load mass is obtained through the load force measuring pin or wire rope tension sensor, and the instantaneous position of the load is obtained through the boom trolley stroke encoder; the central processing unit uses the pulse feedback signal of the tower crane's main hoisting drive and the low-frequency swing data collected by the arm tip inertial measurement unit to verify the tiny swing angles of the load mass in the longitudinal and transverse directions, and thereby correct the instantaneous spatial coordinates of the load center of mass; during the inertia accumulation process, the load mass position and boom boom angle inputs are subjected to exponentially weighted moving average filtering, and abnormal sampling points that exceed the range of three times the median absolute deviation are eliminated through the chi-square test to improve the statistical robustness of the inertia estimation.

[0015] Furthermore, at the end of each control cycle, the central processing unit performs a least squares self-calibration operation by comparing the measured angular acceleration output by the rotational angular velocity sensor with the theoretical angular acceleration predicted based on the real-time inertia model, and makes slight adjustments to the residuals of the boom center of mass position and the load center of mass position in the coordinate transformation matrix; when the numerical change of the real-time equivalent moment of inertia relative to the previous cycle exceeds the safety threshold, the inertia freezing logic is triggered and an inertia abnormality alarm is issued to the tower crane main controller.

[0016] Furthermore, step 3 specifically includes: the central processing unit simultaneously calls the wind-induced coupling torque determined in step 1 and the real-time equivalent moment of inertia determined in step 2 in each control cycle, and calculates the original amplitude torque instruction output by the tower crane amplitude motor based on the established tower crane amplitude dynamics model to actively compensate for high wind load disturbances; the calculation process includes: forming a closed-loop error signal group containing error, error change rate and historical integral according to the target amplitude angle trajectory and the real-time measured amplitude angle; generating an inertia feedforward component with the real-time equivalent moment of inertia and the expected amplitude angle acceleration to offset the uncertainty of mass distribution changes; combining the damping component obtained by filtering the amplitude motor speed, the disturbance compensation amount formed by the projection of the load gravity and the wind-induced coupling torque in the amplitude plane, and performing hierarchical synthesis with the closed-loop error signal group; outputting the original amplitude torque instruction to complete the adaptive control of the amplitude mechanism.

[0017] Furthermore, integral anti-saturation logic is embedded in the original variable amplitude torque command generation stage, and the integral accumulation speed is limited according to preset conditions to prevent strong gusts from causing instantaneous torque overshoot; when the original variable amplitude torque command is lower than the safety lower limit, the original variable amplitude torque command is set to zero to avoid motor jitter in the low load area; when the original variable amplitude torque command is higher than the safety upper limit, the peak is clipped according to a linear decreasing curve and a potential overload risk signal is sent; the original variable amplitude torque command processed by soft limiting is cross-checked with operating parameters such as DC bus voltage, braking resistor temperature rise and servo drive phase margin. If it is found that the braking unit is saturated or the bus voltage exceeds the limit, the energy dissipation priority strategy is triggered, and part of the potential energy feedback sequence is allocated to the rotation or lifting drive to achieve cross-mechanism energy balance.

[0018] Furthermore, in step 4, the central processing unit uses the wind-induced coupling torque obtained in step 1, the real-time equivalent moment of inertia determined in step 2, and the original amplitude-changing torque instruction output in step 3 as input boundary conditions in each control cycle, and performs the following operations according to the tower crane rotation dynamics and energy conservation principle to achieve rotation stability control under high wind load environment: dispatch the projection data of the wind-induced coupling torque in the rotation direction, and perform vector-level synthesis with the overturning component converted from the load gravity in the same cycle to obtain the instantaneous synthetic torque that describes the real-time external disturbance state; call the real-time The instantaneous synthetic torque is converted into the target angular acceleration by using the equivalent moment of inertia, and the actual angular velocity and angular acceleration fed back by the slewing drive are combined to form a closed-loop error signal set including the velocity deviation, acceleration deviation and historical velocity integral. According to the energy shaping idea, the coupling kinetic energy term between the luffing angular velocity and the slewing angular velocity is calculated, mapped into the angular velocity adjustment amount and superimposed on the velocity deviation to form a preliminary corrected angular velocity setting value, so as to update the slewing angular velocity setting value of the tower crane in the next control cycle in real time, thereby completing the linked adaptive control of the luffing-slewing of the tower crane.

[0019] Furthermore, the central processing unit is also equipped with an extreme wind load safety de-control mechanism, which specifically includes: when the transient impact value monitored by the slewing acceleration sensor exceeds the structural limit threshold, it immediately enters a three-level de-control mode, in which the first stage freezes the slewing servo angular velocity command and broadcasts a coordinated deceleration request to the variable amplitude drive module; the second stage reduces the impact of the wind-induced sudden rise on the slewing system by dynamically adjusting the horizontal damping of the hook and the opening of the tower top wind baffle; the third stage gradually releases the freeze and resumes the closed-loop adaptive correction process after confirming that the wind speed and the swing amplitude of the load have fallen back to a controllable range; during the de-control period, the residual wind-induced coupling torque in the slewing direction is continuously calculated in a low-frequency scanning manner, and normal control is automatically re-enabled when the safety window appears.

[0020] The present invention's adaptive control method for tower crane luffing and slewing linkage in high-wind-load environments offers the following advantages: By deploying multi-point ultrasonic wind measurement devices, combined with a coordinate transformation algorithm and real-time attitude information, it achieves high-frequency sensing and spatial decomposition of the instantaneous wind load state, enabling the control system to dynamically and accurately obtain the instantaneous torque of wind-induced disturbances acting on the boom and slewing axis. Based on this information, the central processing unit constructs a comprehensive wind load field and maps it to the slewing axis coordinate system, ensuring high responsiveness and real-time coupling in the calculation of wind-induced coupling torques. By integrating the relationship between load mass, boom attitude, and real-time position, the system performs online identification of load inertia and adaptive correction of boom motion inertia within each control cycle, updating the equivalent moment of inertia model in real time, significantly improving the physical consistency and execution accuracy of torque commands. Furthermore, the present invention incorporates inertia feedforward, damping compensation, and disturbance projection into the luffing torque generation process. By embedding anti-saturation logic and a soft-limiting strategy, it effectively suppresses instantaneous overshoot in strong winds, reduces motor load fluctuations, and improves actuator stability. In addition, during the slewing control stage, through the dynamic integration of the wind load synthetic torque, slewing inertia and kinetic energy coupling, a real-time adaptive correction of the angular velocity set value is achieved, ensuring that the tower crane can maintain a stable posture and accurate hoisting trajectory under sudden changes in wind speed or structural load changes. Especially when extreme wind conditions occur, the present invention has a three-level retreat control process, covering angular velocity freezing, structural energy consumption adjustment and control recovery mechanism, building a global safety control closed loop, and significantly enhancing the fault tolerance and robustness of the system. Overall, the present invention establishes a tightly coupled linkage logic from wind field perception, structural response, dynamic modeling to control output, solving key problems such as model lag, response lag and execution mismatch of traditional control methods in high wind disturbance environments, and improving the safety, accuracy and continuous operation capability of tower cranes under complex construction conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A schematic diagram of a method flow chart of a tower crane luffing-slewing linkage adaptive control method for high wind load environments provided by an embodiment of the present invention;

[0022] Figure 2 A graph showing the test results of the linkage control performance provided by an embodiment of the present invention;

[0023] Figure 3 A comparison diagram of the variable amplitude torque control effect provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0025] refer to Figure 1 : A tower crane luffing-slewing linkage adaptive control method for high wind load environments, the method comprising:

[0026] Step 1: Obtain the measured average wind speed at the tower crane's height. Combined with the crane's slewing angular velocity, boom length, luffing angle, and slewing angle, calculate the boom's windward projected area and relative wind speed to determine the wind-induced coupling torque acting on the crane's slewing axis.

[0027] Step 2: Calculate the vertical projection distance from the tower crane's slewing center to the load based on the crane's inherent moment of inertia, boom deadweight, load mass, boom length, luffing angle, and slewing angle. Calculate the real-time equivalent moment of inertia generated by the boom deadweight, load mass, and their relative positions.

[0028] Step 3: Obtain the original luffing torque command for the tower crane luffing motor output by multiplying the real-time equivalent moment of inertia and the luffing angular acceleration command, the viscous damping effect of the luffing mechanism, the gravitational torque generated by the load, and the projection of the wind-induced coupling torque in the luffing direction;

[0029] Step 4: Using the coupled rotational inertia of the rotating shaft, the projection of the calculated wind-induced coupling torque in the rotation direction, the overturning component of the gravity generated by the hoisted mass on the rotating shaft, and the kinetic energy coupling effect caused by the linkage between the luffing angular velocity and the rotation angular velocity are converted into a correction value for the rotation angular velocity. This is used to update the rotation angular velocity set value of the tower crane in the next control cycle in real time, thereby completing the linked adaptive control of the tower crane's luffing and rotation.

[0030] In a high wind load environment, the essence of the dynamic control problem of a tower crane is: under the dual uncertainties of strong wind disturbances and load changes, how to coordinate the tower crane's luffing mechanism and slewing mechanism, two systems with coupled dynamic behaviors, so that the dynamic response of the entire system is neither unstable nor fast and accurate. Traditional tower crane control systems are mainly based on a separate control architecture, that is, the luffing and slewing mechanisms are designed independently of each other. However, in high wind load scenarios, the two are strongly coupled due to nonlinear aerodynamic coupling and inertial force transmission, which leads to phenomena such as "sway angle amplification", "load oscillation", and "arm tip trajectory deviation" in the system. Therefore, the method proposed in the present invention takes "physical consistency of system modeling" and "online adaptation of dynamic parameters" as the core, and controls the two mechanisms of luffing and slewing as a coupled dynamic system through unified modeling of wind-induced disturbances, mass distribution, driving characteristics and linkage behaviors.

[0031] From a system architecture perspective, its core principle is a four-stage closed-loop chain: disturbance perception - state reconstruction - torque distribution - speed regulation. The first stage is disturbance perception, which essentially constructs a physical picture of the "disturbance field driving quantity" using real-time meteorological sensing systems and boom motion state information. In this stage, wind speed is not an independent variable. Instead, it is coupled with the tower crane's motion (primarily slewing speed and luffing attitude) to form a "relative wind speed field." The aerodynamic projection of this wind speed field on the tower crane's boom is spatially transformed and acts on various controlled objects, such as the slewing and luffing axes. Therefore, wind load cannot be simply viewed as an external input disturbance but must be factored into the state estimation process. The second stage is state reconstruction, which involves the real-time calculation of the dynamic equivalent moment of inertia. This is a key step in achieving the system's "adaptive" nature, as the tower crane's load profile is fluid; its moment of inertia continuously changes due to factors such as the load mass, luffing angle, load swing, and wind-induced positional displacement. If these inertia changes are not sensed and compensated, they will lead to controller parameter mismatch and actuator overshoot or undershoot. To this end, the system needs to build a real-time inertia model based on the decoupling expression of the tower crane's dynamic structure and dynamically adjust it according to the posture and load state. This allows the control algorithm to consider the physical changes of the "controlled object" when distributing torque, thereby improving the robustness of the controller.

[0032] The third stage is torque distribution, which involves mapping system state variables (including attitude, velocity, acceleration, and inertia) along with external disturbances (wind force and gravity torque) to drive commands, enabling adaptive control of the luffing motor's output torque. The core technical principle of this stage is the "physics-consistent control law." The controller no longer relies on error feedback based on a fixed model (as in traditional PID control), but instead solves a mechanical equilibrium based on the system's inertial characteristics and disturbance structure. In other words, each frame of control command reflects the system's physical reality, rather than empirical adjustments or fixed gains. In this control structure, torque generation is not a response to error, but rather the actual force required to achieve the desired trajectory. The fourth stage is speed regulation, primarily targeting the slewing mechanism. This component is coupled with the luffing mechanism. In particular, when the luffing acceleration changes or the wind load direction shifts, the slewing axis experiences coupled inertial forces and aerodynamic torques. The principle underlying this stage is "momentum energy consistency adjustment," which means using the equivalent moment of inertia as a reference to convert the total disturbance torque (wind force, gravity component, and amplitude linkage term) acting on the rotating axis into a change in angular acceleration, which is then used to adjust the set value of the rotational angular velocity for the next cycle. This adjustment is not based on speed limitations imposed by a static model, but rather on a dynamic reconfiguration of the momentum transfer mechanism at the current moment. In essence, this constitutes a small-step angular momentum conservation prediction mechanism, the purpose of which is to avoid the "secondary sway" caused by inertial coupling while maintaining control accuracy.

[0033] The key to achieving coordinated control in this method lies not in simply connecting the slewing and luffing controllers in series, but rather in incorporating a mechanism called "state-to-state modeling" into the control logic. Specifically, the momentum contribution of the luffing state is considered in slewing control, while the influence of the slewing speed on the relative wind speed is considered in the luffing torque. This effectively couples the two previously separately controlled systems through the state propagation chain, transforming the controller design into a unified multivariable closed-loop system. This method is particularly suitable for high wind load environments. At low wind speeds, aerodynamic forces contribute minimally to the overall system energy structure and can be considered small perturbations, allowing the slewing-luffing coupling term to be negligible. However, at wind speeds exceeding 15 meters per second or even 20 meters per second, the rate of change of wind-induced torque exceeds the inertial response speed. Failure to consider the combined effects of relative wind speed, structural attitude, and inertia distribution can lead to system overshoot, frequent braking, and inability to track the trajectory, ultimately resulting in excessive load swing or even loss of control. This invention incorporates the active dynamic characteristics of wind-induced disturbances into its modeling framework, forming a closed loop from system identification to drive control, thereby achieving comprehensive management of high-wind disturbances through the "absorption-distribution-coordination" process. Furthermore, the adaptability of this method is not driven solely by neural networks or black-box algorithms, but rather through analytical closed-loop self-regulation achieved through real-time parameter modeling.

[0034] Furthermore, in step 1, ultrasonic wind measuring devices are respectively arranged near the tip of the tower crane boom, the tower cap position, and at predetermined heights of multiple layers above the tower base to synchronously collect the incoming wind speed, wind direction, and turbulent pulsation intensity as raw meteorological data; the raw meteorological data is timestamped according to a unified clock and sent to the central processing unit to form a wind field raw data set covering all measuring points within each control cycle.

[0035] To achieve time synchronization, each wind measuring device uses a unified time reference through a time synchronization protocol (such as IEEE1588 or an internal high-precision clock synchronization module). This ensures that within any control cycle, the data collected at each measuring point has a strictly consistent timestamp, eliminating the timing deviation that may occur during spatial data integration. Through this structured time control, the raw meteorological data is integrated into a single wind field raw data set according to a unified time reference when it is sent to the central processing unit. This data set not only contains the wind speed value, wind direction angle, and turbulence index at each sampling point at the current moment, but also reflects the spatial gradient distribution of the local wind field through the spatial distribution relationship between the measuring points.

[0036] After receiving the raw wind farm data set, the central processing unit further integrates it with the crane's structural parameters and current attitude state to construct a "crane-coupled wind disturbance field model." In this model, the wind speed data at each measurement point is no longer an isolated sample. Instead, it is mapped to the crane's reference coordinate system and correlated with the boom's attitude in real time to obtain the "position change response of the wind load direction relative to the crane's coordinate system." This model is a key innovation that distinguishes this control method from traditional wind speed threshold speed limiting strategies. It can capture the spatial dynamic interaction between crane motion and wind speed changes, thereby enabling dynamic prediction of subsequent wind-induced coupling torques.

[0037] In particular, the turbulence fluctuation intensity, a key parameter characterizing the frequency characteristics of wind speed disturbances, is not simplified to a mean wind speed deviation in this step. Instead, it is fully retained as an input variable, participating in the construction of the disturbance term of the subsequent wind torque rapid change prediction model. This allows for early compensation of the rising edge response of wind-induced disturbances in the event of sudden high wind load changes. Ultimately, this structurally complete, temporally consistent, and spatially clearly distributed wind farm raw data set serves as the basis for physical modeling, providing data support for subsequent wind-induced coupled torque calculations, disturbance projection, and linkage control. It is also input into the main control loop in the form of a structured cache, ensuring that the tower crane control system has the ability to continuously update, quickly respond, and dynamically adapt to high wind load disturbances.

[0038] Furthermore, the central processing unit performs the following processes: multi-stage filtering is performed on the original data set of the wind farm, specifically including: using bandpass filtering to suppress low-frequency drift and high-frequency noise, using a sliding time window to smooth gust pulsation, and removing instantaneous distortion samples according to a preset abnormality detection threshold to obtain effective wind speed and effective wind direction with stable statistical characteristics; according to the real-time amplitude angle, rotation angle and boom length of the tower crane, the coordinate transformation algorithm is called to decompose the effective wind speed into a tangential component consistent with the longitudinal direction of the boom and a normal component perpendicular to the boom plane; according to the windward projection area data of the boom The equivalent wind-exposed area under different postures is matched by table lookup, and the tangential component and the normal component are combined with the equivalent wind-exposed area to construct a comprehensive wind load field reflecting the instantaneous aerodynamic state of the tower crane; the comprehensive wind load field is mapped to the slewing axis coordinate system, and the wind-induced coupling torque acting on the slewing axis is obtained by the integration method according to the length of the effective arm from the boom to the slewing center and the position of the center of mass of the boom; the slewing angular velocity and the tangential linear velocity of the boom tip are read synchronously, and the wind-induced coupling torque is coupled and corrected in combination with the real-time wind direction deviation angle to reflect the real-time influence of the slewing motion on the relative wind speed.

[0039] Real-time wind-induced coupling torque for:

[0040] ;

[0041] in, is the air density; is the arm section resistance coefficient; is the windward projected area of ​​the boom; is the measured average wind speed at the sampling height; is the current rotation angular velocity; is the boom length; is the amplitude angle; is the rotation angle.

[0042] Under high wind loads, the structural stress characteristics and the nature of the control object of the tower crane undergo fundamental changes. Wind is no longer a simple background disturbance, but a dynamically dominant external energy input source. Its scope of action not only includes the load and boom, but also directly couples into the tower crane's rotational dynamics and amplitude response, thereby changing the state evolution of the entire system. In order to make the tower crane's control system capable of active wind resistance, the first issue that must be addressed is how to accurately quantify the coupled effect of wind on the structure using controllable logic and unify this quantity with the structural state in the control model. The design of Step 1 is a systematic solution to this problem.

[0043] Its core principle is to convert wind field disturbance information into equivalent dynamic load descriptors in the structural reference frame. This allows wind to serve as a "driving input" and enter the system dynamics model along with intrinsic parameters such as mass, inertia, position, and velocity, thereby becoming part of the feedback control. This conversion does not directly impose control limits or simple thresholds on wind speed. Instead, it treats wind speed as a dynamic field variable, decouples it from its attitude through spatial projection, and maps it into "control influencing parameters," such as coupled wind torque. This principle is premised on the fact that tower cranes are not point-mass systems, but rather nonlinear systems with high-order structural distributions and three-dimensional rigid-body properties. The influence of wind speed on them cannot be described simply as "wind speed values" but must be converted into forces and torques acting on the structural nodes to contribute to the derivation of the system's control laws. These forces and torques depend not only on the wind speed itself but also on the combined effects of the wind speed and the relative direction of the structure, the windward projection of the structural surface, and the velocity field of the structure during motion.

[0044] Therefore, the rationale behind setting up multiple wind measurement points in step 1 is not to achieve "multi-point monitoring for greater accuracy," but rather to obtain spatially accurate information on the local wind field gradient distribution, thereby reconstructing the non-uniform effects of the wind speed field on various parts of the tower crane structure. During luffing and slewing, the tower crane's posture constantly changes, and the relationship between the wind's angle of impact and the structure's projected surface adjusts accordingly. Using only a single wind speed value or a simple estimation method would lose the control system's ability to characterize the geometric coupling characteristics of the disturbance source, inevitably leading to delayed or even erroneous control output. By constructing a wind speed field using multi-point data, the system can "redefine" wind speed as a function of the relative wind field in the structural reference frame based on the tower crane's inherent motion state, rather than as a single external disturbance. Based on this foundation, the further processing in step 1—projection of wind speed onto the tangential and normal components, matching and table lookup of the projected areas, and synthesis of the wind load field—essentially constructs a "wind-induced kinetic energy input model." This model is not used to directly control wind speed, but rather to estimate the spatially directional disturbance torque exerted by the wind on the tower crane, which is then used as part of the system state in the subsequent control law generation. The emergence of wind-induced coupling torque is not a result of artificial design but rather the result of the introduction of field variables into structural control through system modeling. It is an asymmetric disturbance that dynamically adjusts the objectives of the slewing and luffing control strategies without changing the crane structure. This shifts the control system from a simple "trajectory tracking" logic to a "dynamic regulation under state-disturbance coupling" logic.

[0045] Furthermore, step 2 specifically includes: reading the inherent moment of inertia of the tower body and slewing platform, the deadweight of the boom, the length of the boom, and the geometric position of the center of mass of the boom and the center of rotation as a set of static design parameters, and locking the static design parameter set once based on the factory calibration report; obtaining the real-time value of the load mass and the instantaneous position of the load in the longitudinal direction of the boom, combining the amplitude angle and rotation angle information shared in step 1, and performing coordinate transformation on the vertical projection distance of the load in the rotation plane and the action arm from the boom to the center of rotation to construct a dynamic position model of the load mass in the rotation axis coordinate system; according to the dynamic equivalence principle of the tower crane, the inherent moment of inertia of the tower body and slewing platform, the deadweight of the boom and its inertia component caused by the change of the amplitude angle, and the inertia contribution of the load mass to the rotation axis at the real-time position are accumulated separately, and the coupling inertia cross term between the boom and the center of mass of the load relative to the rotation center is introduced in the accumulation process to obtain the real-time equivalent moment of inertia.

[0046] Real-time equivalent moment of inertia for:

[0047] ;

[0048] in, is the inherent moment of inertia of the tower body and the slewing platform; The arm is deadweight; is the mass of the lifting load; It is the vertical projection distance from the center of rotation to the load.

[0049] In practice, the process begins with reading the inherent moment of inertia of the tower body and slewing platform, the deadweight of the boom, the boom length, and the geometric position of the boom's center of mass and center of slew as input. These parameters constitute a set of static design parameters for the tower crane structure. These parameters were obtained during the crane's manufacturing phase through finite element analysis, center of gravity calibration, and whole-machine counterweight testing. After installation and commissioning, they are solidified in the form of a factory calibration report. The central processing unit reads and locks these parameters once during system initialization, and they remain unchanged during subsequent operation, ensuring the consistency and traceability of the system control model. These static design parameters not only provide a baseline inertia term for subsequent calculations but also constitute the immutable input portion of the inertia model that is directly related to the structural configuration.

[0050] To incorporate dynamics, the system also requires real-time variables related to the operating state. First, a force measuring device installed on the hook or hoisting wire rope tension path acquires the real-time value of the load mass. Furthermore, since the load's position along the boom is not fixed and its effect on the slew center varies at different lifting heights, the system simultaneously acquires the load's instantaneous position along the boom's longitudinal direction. This is typically derived indirectly from the position of the luffing trolley or the change in the length of the hoisting rope. Based on the load mass and position, the central processing unit (CPU) uses the luffing angle and slew angle information shared in step 1 to map the load's displacement relative to the boom into a spatial coordinate system within the crane's slewing plane. This constructs the load's vertical projection within the slewing plane and further calculates the load's geometric arm relative to the slew center through a coordinate transformation. This transformation converts the actual spatial displacement into a dynamic projection within the slewing axis reference coordinate system. The result is used to assess the degree of inertia exerted by the load mass on the slewing axis.

[0051] After completing the static and dynamic parameter collection, the central processing unit begins executing a real-time inertia solution based on the tower crane dynamics equivalence principle. During this process, the system first introduces the inherent rotational inertia of the tower body and slewing platform into the calculation model as a constant term. It then processes the boom's deadweight. Considering that the variable amplitude angle causes the boom's center of gravity to change in spatial position relative to the slewing axis, the system adjusts the distance from the boom's center of gravity to the slewing center in real time based on the amplitude angle state, thereby correcting the boom's deadweight contribution to the rotational inertia. The system then independently estimates the effect of the load mass on the slewing axis's inertia at the current instantaneous position. This effect is not only related to the load mass and the distance to the lifting point, but also affected by the boom's posture on the projected length of the moment arm. Therefore, the system needs to consider the three-dimensional angle between the load point and the slewing center in real time during the calculation.

[0052] After completing the estimation of all independent components, the central processing unit accumulates them item by item to construct a complete basic inertia synthesis expression. However, simply accumulating the individual inertias cannot fully reflect the coupling behavior of the actual system. In the state of variable amplitude-rotation linkage, the arm's deadweight and the load mass will form a linkage inertia effect during the movement, resulting in the appearance of mutual cross-influence terms in the actual rotational inertia. To this end, the system introduces a coupling inertia cross term between the arm's center of mass and the load's center of mass relative to the center of rotation, and embeds this term as a linkage compensation in the inertia model during the construction of the real-time equivalent rotational inertia. This cross term is derived from the structural kinematic expression and reflects the influence of the two mass bodies on the total angular momentum response in the non-collinear motion state. It is one of the most core nonlinear terms in the entire linkage control modeling.

[0053] Furthermore, the real-time value of the load mass is obtained through the load force measuring pin or wire rope tension sensor, and the instantaneous position of the load is obtained through the boom trolley stroke encoder; the central processing unit uses the pulse feedback signal of the tower crane's main hoisting drive and the low-frequency swing data collected by the arm tip inertial measurement unit to verify the tiny swing angles of the load mass in the longitudinal and transverse directions, and thereby correct the instantaneous spatial coordinates of the load center of mass; during the inertia accumulation process, the load mass position and boom boom angle inputs are subjected to exponentially weighted moving average filtering, and abnormal sampling points that exceed the range of three times the median absolute deviation are eliminated through the chi-square test to improve the statistical robustness of the inertia estimation.

[0054] Furthermore, at the end of each control cycle, the central processing unit performs a least squares self-calibration operation by comparing the measured angular acceleration output by the rotational angular velocity sensor with the theoretical angular acceleration predicted based on the real-time inertia model, and makes slight adjustments to the residuals of the boom center of mass position and the load center of mass position in the coordinate transformation matrix; when the numerical change of the real-time equivalent moment of inertia relative to the previous cycle exceeds the safety threshold, the inertia freezing logic is triggered and an inertia abnormality alarm is issued to the tower crane main controller.

[0055] In the adaptive control method for tower crane luffing and slewing linkage in high wind load environments, the central processing unit (CPU) not only calculates the equivalent moment of inertia in real time but also must ensure the credibility, convergence, and physical consistency of this inertia parameter in a dynamic environment. Because the calculation of the equivalent moment of inertia is highly dependent on the spatial positional estimates of the boom and load centers of mass, which can experience time-varying deviations due to factors such as sensor errors, mechanical backlash, and wind-induced deformation, relying solely on forward extrapolation will cause the inertia model to gradually deviate from its true physical value after multiple cycles of operation. To ensure that the inertia parameters maintain physical consistency during control, the CPU has designed a least-squares self-calibration mechanism driven by sensor feedback.

[0056] The core principle of this self-calibration mechanism is to compare the angular acceleration measured by the slewing angular velocity sensor as the actual physical response with the theoretical angular acceleration derived from a real-time equivalent moment of inertia model constructed during the current cycle and the known resultant torque. The theoretical angular acceleration, derived from the system dynamics model combined with real-time input, represents the expected system response under ideal modeling assumptions. The angular acceleration provided by the sensor, on the other hand, represents the feedback output of the tower crane under the combined effects of wind loads, inertia, and structural deviations in a real-world environment. Systematic deviations between the two indicate an error source in the modeling process, particularly a static offset or dynamic drift in the positions of the boom and load centers of mass during the coordinate transformation. The system then invokes a least-squares self-calibration algorithm, using the error between the measured and predicted angular accelerations as the objective function. The algorithm minimizes the sum of squared residuals over several consecutive control cycles, thereby inversely adjusting the relative position offset parameters of the boom and load centers of mass in the coordinate transformation matrix to achieve a subtle correction to the inertia estimation model.

[0057] This correction process does not aim to remodel the system. Instead, it fine-tunes the geometric parameters of the existing model online, achieving dynamic compensation within the control frequency range. This results in a system inertia model with "finite convergence and gradual correction" performance. The corrected results are written back to the system status register at the end of the control cycle and serve as the initial parameter input for the inertia calculation in the next cycle, achieving iterative model convergence.

[0058] To ensure system stability and avoid model oscillation, the central processing unit also incorporates an inertia change monitoring mechanism, which compares the change in the equivalent moment of inertia calculated for the current cycle with the inertia value from the previous cycle in real time. When the change exceeds a safety threshold defined by the manufacturing tolerances of the tower crane structure and the physical limits of load variation, the system determines that the current inertia model may have significant uncertainty or be affected by non-physical interference. At this point, the inertia freeze logic is immediately triggered, forcing the inertia value of the current cycle to remain at the stable value of the previous cycle, prohibiting further updates and recording this status as an inertia anomaly event. An inertia anomaly alarm is simultaneously issued to the tower crane main controller, alerting the operation and maintenance system or operator that the current state may be due to sensor failure, severe wind load disturbances, uncontrolled load swing, or other unexpected structural responses, thereby preventing unreliable inertia data from causing over- or under-adjustment of control commands.

[0059] While the inertia freeze state persists, the central processing unit retains the least squares error monitoring function, but does not immediately update the model parameters. Instead, it enters observation mode, continuously recording the residual sequence between the measured angular acceleration and the frozen model prediction. After the residual converges back to the safe error band over multiple consecutive cycles, the freeze state is automatically released, and the self-calibration function and inertia update process are restored, ensuring that the control system has online self-recovery and dynamic adaptability while ensuring safety. This mechanism ensures the physical validity of the core control parameters and prevents non-substantial drift of the inertia model under multi-source disturbances. It is the key supporting mechanism for maintaining model credibility and closed-loop stability in the linkage control structure of the present invention.

[0060] Furthermore, step 3 specifically includes: the central processing unit simultaneously calls the wind-induced coupling torque determined in step 1 and the real-time equivalent moment of inertia determined in step 2 in each control cycle, and calculates the original amplitude torque instruction output by the tower crane amplitude motor based on the established tower crane amplitude dynamics model to actively compensate for high wind load disturbances; the calculation process includes: forming a closed-loop error signal group containing error, error change rate and historical integral according to the target amplitude angle trajectory and the real-time measured amplitude angle; generating an inertia feedforward component with the real-time equivalent moment of inertia and the expected amplitude angle acceleration to offset the uncertainty of mass distribution changes; combining the damping component obtained by filtering the amplitude motor speed, the disturbance compensation amount formed by the projection of the load gravity and the wind-induced coupling torque in the amplitude plane, and performing hierarchical synthesis with the closed-loop error signal group; outputting the original amplitude torque instruction to complete the adaptive control of the amplitude mechanism.

[0061] The original variable amplitude torque command is :

[0062] ;

[0063] in, is the variable angular acceleration instruction; is the equivalent viscous damping coefficient of the luffing mechanism; is the amplitude angular velocity; is the acceleration due to gravity.

[0064] In the adaptive control method for tower crane luffing and slewing linkage in high wind load environments, adaptive control of the luffing mechanism is a core component of the entire system's linkage logic. Its role is not only to achieve basic adjustments to the load height and operating radius, but more importantly, to actively compensate for the structural response caused by wind-induced disturbances, ensuring consistency between the tower crane's boom tip trajectory and the mission planning curve, and minimizing secondary load sway. To achieve this control objective, the central processing unit dynamically generates the output torque command for the luffing motor within each control cycle based on the tower crane's luffing dynamics model, integrating the wind-induced coupling torque from step 1 with the real-time equivalent moment of inertia from step 2.

[0065] First, the system performs real-time closed-loop error analysis on the amplitude angle trajectory. This analysis uses the target amplitude angle trajectory provided by the upper-level task planning module as a reference, compares it with the amplitude angle displacement measured in the current cycle, and calculates the instantaneous position error. The error rate of change is obtained based on the first-order difference of the position error. The system also maintains an integral cumulative value within a time window to form a closed-loop error signal group. This signal group is the core feedback quantity of the controller, reflecting the current system's attitude deviation and its dynamic change trend. Compared with proportional control using only position error, this signal group has stronger disturbance response capability and better steady-state tracking performance.

[0066] Based on the error signal set, the system introduces the product of the real-time equivalent moment of inertia and the desired variable angular acceleration as a feedforward compensation term. This term reflects the system's predictive response to dynamic load inertia changes. It can effectively offset output hysteresis or error accumulation caused by structural inertia changes, especially when the load mass changes rapidly or the boom accelerates significantly during extension and retraction. This feedforward compensation is based on the real-time equivalent moment of inertia calculated in step 2 of the current cycle. It is independent of fixed model parameters and has adaptive adjustment capabilities. It is a key mechanism for the system to actively adapt to changes in structural inertia.

[0067] The central processing unit then introduces a damping component based on the luffing motor's operating status. This component is derived from the luffing motor's current speed signal. After digital filtering, a smooth velocity estimate is generated, which is used to construct a virtual damping torque proportional to the speed. This damping term primarily suppresses residual vibrations generated during rapid boom movement. This is particularly true under high wind loads, as the wind speed disturbance spectrum contains high-frequency components, requiring a strong speed response. Therefore, introducing the damping term effectively reduces system oscillations and improves controller stability.

[0068] To further enhance the controller's adaptability to external disturbances, the system combines the component of the wind-induced coupling torque in the luffing direction from step 1 with the equivalent torque exerted on the luffing mechanism by the load's gravity in the current attitude to form a disturbance compensation. This disturbance compensation is a real-time estimate of the system's response to both wind and gravity disturbances. It allows the controller to preload torque before significant position errors occur, thus achieving a concurrent "disturbance feedforward + attitude feedback" control strategy, enhancing the control system's disturbance transparency and proactive response.

[0069] After all the aforementioned components are constructed, the central processing unit performs a weighted fusion of the feedforward component, damping component, disturbance compensation, and closed-loop error signal group according to the preset hierarchical synthesis logic to generate the original variable amplitude torque command. The hierarchical synthesis strategy reflects the system's priority management of different control objectives, namely, emphasizing error integration and damping stability in steady-state control, and feedforward accuracy and disturbance suppression capabilities in dynamic transitions. The resulting original variable amplitude torque command is the result of multi-source information fusion, possessing temporal consistency, physical consistency, and disturbance resistance, and can be directly sent to the servo drive for variable amplitude motor control execution.

[0070] Through this control process, the system can not only maintain the accuracy of the boom trajectory under high wind load environments, but also effectively suppress the dynamic instability of the luffing system caused by sudden changes in wind speed, load changes, and structural inertia drift. It provides stable and reliable dynamic boundary conditions for the slewing control of the entire tower crane, and constitutes an important executive support unit in the luffing-slewing linkage control strategy.

[0071] Furthermore, integral anti-saturation logic is embedded in the original variable amplitude torque command generation stage, and the integral accumulation speed is limited according to preset conditions to prevent strong gusts from causing instantaneous torque overshoot; when the original variable amplitude torque command is lower than the safety lower limit, the original variable amplitude torque command is set to zero to avoid motor jitter in the low load area; when the original variable amplitude torque command is higher than the safety upper limit, the peak is clipped according to a linear decreasing curve and a potential overload risk signal is sent; the original variable amplitude torque command processed by soft limiting is cross-checked with operating parameters such as DC bus voltage, braking resistor temperature rise and servo drive phase margin. If it is found that the braking unit is saturated or the bus voltage exceeds the limit, the energy dissipation priority strategy is triggered, and part of the potential energy feedback sequence is allocated to the rotation or lifting drive to achieve cross-mechanism energy balance.

[0072] In the adaptive control method for tower crane luffing and slewing linkage in high wind load environments, the generation of the original luffing torque command is not only a calculation process based on error feedback and disturbance compensation, but also a dynamic output control process involving multi-level safety mechanisms and energy coordination strategies. Because wind-induced disturbances under high wind load conditions are often sudden and high-amplitude, the lack of restrictive management of control commands can easily lead to a series of system instability issues such as torque command mutations, drive overload, structural impact, or vibration amplification. Therefore, when generating the original luffing torque command, the central processing unit embeds an integrated set of integral anti-saturation logic and operating state-related limiting mechanisms to ensure the physical executable of the control output and the electrical safety of the system operation.

[0073] First, integral anti-windup logic is embedded in the processing path of the closed-loop error signal group. Its control concept is based on setting the integral speed adjustment condition to dynamically limit the accumulation rate of the error integral. When the amplitude error is detected to be in a continuous rising state, or the error change rate is increasing exponentially, and the system is currently in a high wind speed disturbance period, the system will automatically reduce the response speed of the integrator or directly freeze the integral path when necessary, thereby avoiding a sudden surge in the control torque caused by the error integral. This mechanism effectively curbs the amplification effect of strong gusts or local turbulence on the control output and prevents system instability caused by integral overshoot in the torque response.

[0074] Subsequently, the central processing unit sets a dual soft limit mechanism for the numerical range of the original variable amplitude torque instruction, and makes judgments and executions based on the actual operating conditions. If the calculated original variable amplitude torque instruction is lower than the preset safety lower limit, the system will directly set the instruction to zero, thereby avoiding periodic jitter or oscillation when the motor is continuously running in the extremely low load area, and ensuring the stability of the servo actuator. On the contrary, if the original variable amplitude torque instruction is higher than the safety upper limit, the system does not directly cut it, but gradually cuts the peak according to the linear decreasing curve, so that the control output presents a predictable and continuous downward adjustment process when approaching the execution limit, reducing the transient impact of the torque boundary effect on the actuator. At the same time, when the high limit is cut, the central processing unit will also generate and send a potential overload risk signal to the system monitoring module to record the current control status and the safety threshold events that may be triggered, providing a basis for subsequent data audits and abnormal working condition reproduction on the operation and maintenance side.

[0075] After the above-mentioned soft limiting processing, the central processing unit cross-checks the original variable amplitude torque command to be issued in the current cycle with the key boundary parameters of the system operation, including but not limited to the DC bus voltage, the temperature rise level of the brake resistor, and the phase margin of the servo drive. The purpose of the cross-check is to evaluate the feasibility of the execution of the current control command and whether the electrical system has sufficient energy buffering capacity. In actual operation, if the bus voltage approaches the equipment limit, the resistor temperature continues to rise, or the phase margin is close to the system critical point, it means that the current electrical system is in a high-load operation state. Continuing to maintain or increase the torque command may cause system overload or even failure.

[0076] To this end, when the system detects that the operating parameters are out of limit or approaching a critical state during cross-checking, it will trigger an energy dissipation priority strategy. The core idea of ​​this strategy is to guide part of the mechanical energy that cannot be fully dissipated by the variable amplitude mechanism due to the accumulation of braking torque to be transferred to the slewing or lifting drive unit through cross-mechanism coordinated scheduling, so as to share the energy pressure and maintain the overall energy consumption balance of the system. The implementation of this strategy depends on the energy feedback scheduling module within the system. The module determines the energy flow path based on the current load conditions and operating capabilities of each actuator, and issues the adjustable energy value and target allocation mechanism in a non-disturbance manner. In this way, the system not only avoids the torque limitation caused by the energy saturation of a single mechanism, but also realizes the coordinated regulation and energy self-balancing across actuator units under severe disturbances caused by high wind loads.

[0077] Furthermore, in step 4, the central processing unit uses the wind-induced coupling torque obtained in step 1, the real-time equivalent moment of inertia determined in step 2, and the original amplitude-changing torque instruction output in step 3 as input boundary conditions in each control cycle, and performs the following operations according to the tower crane rotation dynamics and energy conservation principle to achieve rotation stability control under high wind load environment: dispatch the projection data of the wind-induced coupling torque in the rotation direction, and perform vector-level synthesis with the overturning component converted from the load gravity in the same cycle to obtain the instantaneous synthetic torque that describes the real-time external disturbance state; call the real-time The instantaneous synthetic torque is converted into the target angular acceleration by using the equivalent moment of inertia, and the actual angular velocity and angular acceleration fed back by the slewing drive are combined to form a closed-loop error signal set including the velocity deviation, acceleration deviation and historical velocity integral. According to the energy shaping idea, the coupling kinetic energy term between the luffing angular velocity and the slewing angular velocity is calculated, mapped into the angular velocity adjustment amount and superimposed on the velocity deviation to form a preliminary corrected angular velocity setting value, so as to update the slewing angular velocity setting value of the tower crane in the next control cycle in real time, thereby completing the linked adaptive control of the luffing-slewing of the tower crane.

[0078] Rotational angular velocity setting value for the next control cycle for:

[0079] ;

[0080] In the tower crane luffing-slewing linkage adaptive control method for high wind load environments, the angular velocity control of the slewing system is not only related to operational accuracy and motion smoothness, but is also a key link in structural safety and load sway control. Under high wind load conditions, due to the non-stationary, spatially non-uniform, and attitude-dependent nature of wind-induced disturbances, angular velocity settings based solely on traditional static models can no longer meet the dynamic control requirements under wind field coupling. Therefore, step 4 introduces an angular velocity adaptive correction mechanism driven by multi-source boundary conditions and capable of coupling kinetic energy compensation. The central processing unit uses the wind-induced disturbance, structural inertia state, and luffing drive output as input variables, and constructs a slewing angular velocity setting value generation path that adapts to wind field changes based on tower crane slewing dynamics and energy conservation principles.

[0081] The process first dispatches the component in the rotation direction from the wind-induced coupling torque obtained in step 1. This is the main source of the disturbance torque directly generated by the wind load on the rotary system. This torque component reflects the relationship between the windward direction and the structural projection in the current posture of the tower crane and is the main driving term of external disturbance in structural motion control. At the same time, the system synchronously reads the mass information of the load in the current cycle, and combines the boom amplitude angle and the vertical projection distance of the load in the rotation plane to calculate the overturning moment component generated by the load gravity on the rotary system. The wind-induced disturbance torque and overturning moment are not linearly superimposed in physical space, but are synthesized in a vector manner in the control model. The resulting instantaneous synthetic torque truly reflects the force distribution of the rotary system under the current structural and load conditions and is the core external input variable used to derive the angular velocity response in this control cycle.

[0082] After obtaining the synthetic torque, the central processing unit calls the real-time equivalent moment of inertia provided in step 2 and maps the disturbance torque to the target angular acceleration based on the dynamic relationship of the tower crane's rotary motion. This angular acceleration represents the response trend that the rotary system should have under the current disturbance state. After that, the system retrieves the actual angular velocity and angular acceleration measurement values ​​fed back by the rotary servo drive in the current cycle, and constructs a closed-loop error signal set containing velocity deviation, acceleration deviation, and historical velocity integral. This error signal set is the core basis for the controller to adjust the angular velocity. It has dynamic response capabilities and cumulative error correction capabilities, and can fully reflect the difference between the current control output of the system and the actual state.

[0083] While forming the error signal set, the central processing unit calculates the kinetic energy coupling relationship between the amplitude modulation motion and the rotational motion based on the system's energy shaping concept. This kinetic energy coupling term originates from the redistribution effect of the amplitude modulation acceleration caused by the amplitude modulation torque instruction in step 3 on the kinetic energy distribution of the entire system. Since the amplitude modulation and the rotation mechanism have spatial cross-actions, their instantaneous velocity vectors have inner product terms in the inertial reference system, resulting in the appearance of non-independent terms in the kinetic energy expression. Therefore, under high wind loads, changes in the amplitude modulation angular velocity will have an indirect impact on the rotation system, manifesting as additional kinetic energy input or output. The system calculates the coupled kinetic energy term through the rate of change between the amplitude modulation angular velocity and the rotational angular velocity in the current cycle, and converts it into an equivalent angular velocity adjustment amount as an additional compensation term for correcting the rotational angular velocity set value.

[0084] This kinetic energy compensation term is superimposed on the velocity deviation in the aforementioned closed-loop error signal set to form a preliminary corrected angular velocity setpoint. This setpoint is no longer a target quantity derived solely from angular error or disturbance torque, but rather a dynamic control output that integrates multiple factors, including structural state, disturbance dynamics, linkage feedback, and energy coupling. It exhibits strong environmental adaptability and proactive response. At the end of the cycle, this corrected angular velocity setpoint is written into the target value register for the next control cycle, used to drive the rotary servo system for updated control and serves as a reference for subsequent control logic execution.

[0085] Furthermore, the central processing unit is also equipped with an extreme wind load safety de-control mechanism, which specifically includes: when the transient impact value monitored by the slewing acceleration sensor exceeds the structural limit threshold, it immediately enters a three-level de-control mode, in which the first stage freezes the slewing servo angular velocity command and broadcasts a coordinated deceleration request to the variable amplitude drive module; the second stage reduces the impact of the wind-induced sudden rise on the slewing system by dynamically adjusting the horizontal damping of the hook and the opening of the tower top wind baffle; the third stage gradually releases the freeze and resumes the closed-loop adaptive correction process after confirming that the wind speed and the swing amplitude of the load have fallen back to a controllable range; during the de-control period, the residual wind-induced coupling torque in the slewing direction is continuously calculated in a low-frequency scanning manner, and normal control is automatically re-enabled when the safety window appears.

[0086] In the adaptive control method for tower crane luffing and slewing linkage designed for high wind load environments, the central processing unit (CPU), in addition to high-frequency closed-loop control capabilities, must also be equipped with an extreme operating condition response mechanism with structural safety assurance to cope with impact load events caused by sudden wind loads. During actual tower crane operation, wind field fluctuations often include low-frequency shear layers, gust fluctuations, and turbulence spikes. Especially under high wind speed boundary conditions, wind-induced loads can rise dramatically in a very short period of time, causing transient shocks to the slewing system that far exceed the inertial response capacity. If such shocks exceed the structural design limits while conventional control logic is maintained, irreversible consequences such as amplified load swing, servo instability, and even structural damage may occur. To this end, the CPU incorporates a three-level progressive extreme wind load safety fallback mechanism. This ensures that the system automatically interrupts the conventional control path upon detecting signs of potential structural overload and implements control degradation and coordinated load reduction strategies based on the principle of safety first.

[0087] The activation conditions of the decontrol mechanism are based on the measured data of the slewing acceleration sensor. The central processing unit continuously monitors the angular acceleration response curve of the slewing system in each control cycle and compares its measured peak value with the limit threshold predefined in the structural database. When the absolute value or rate of change of the angular acceleration exceeds the structural limit threshold, the system immediately enters the first stage decontrol mode. The action of the first stage is to freeze the slewing servo angular velocity instruction, that is, immediately terminate the slewing velocity target update generated in the current cycle, and keep the output state of the servo driver at the current value or safe stagnation state to prevent the system from continuing to advance the attitude correction process during the impact. At the same time, the central processing unit sends a broadcast-style coordinated deceleration request to the amplitude drive module, instructing it to reduce the amplitude angular velocity or actively retract the boom length to reduce the wind load projection area of ​​the entire system, thereby reducing the total structural force in a linkage manner.

[0088] After entering the second phase, the system is no longer limited to the control level, but instead links actuators and auxiliary mechanical devices to carry out wind load reduction operations. First, the response parameters of the hook's horizontal damping module are dynamically adjusted, such as increasing the damper oil circuit throttling ratio or activating the bypass flow limiting channel to enhance the energy dissipation capacity of the load sway. Second, the opening of the tower top wind damper is adjusted according to the direction and amplitude of the wind field. Through aerodynamic means, the wind flow is deflected or decompressed in local areas, reducing the accumulation of wind pressure peaks on the structural surface, thereby reducing the rate of increase of wind-induced coupling torque. This stage focuses on delaying the shock transmission rate through structural-level flexible buffering when external disturbances have not yet been fully released, providing a transition zone for subsequent control recovery.

[0089] After the two stages of response and relief described above, the system enters the third stage, which is to determine whether the conditions for restoring normal control are met. The central processing unit continuously reads the data streams from the wind speed monitoring device and the load swing amplitude estimation module. When the wind speed drops back to the designed safety envelope range and the load swing amplitude is within the controllable range, the system gradually releases the freeze logic, re-enables the real-time update function of the rotational angular velocity setting value, and resumes the adaptive correction process. This release process is not an instantaneous recovery, but rather a transition to the target angular velocity setting state through smooth interpolation within multiple cycles, ensuring continuity and response stability when the system re-enters the closed-loop control path, avoiding the "recovery jump" phenomenon.

[0090] Throughout the decontrol cycle, despite the frozen angular velocity command, the central processing unit maintains its low-frequency scanning mechanism, periodically calculating the residual component of the current wind-induced coupling torque in the rotational direction and assessing the stability of wind farm recovery based on its changing trend. When the residual wind-induced torque significantly decreases or stabilizes, and the fluctuation amplitude is within a preset range, the system deems that a "safe window" exists for re-entering normal control. The system automatically determines this and enters a control process reset state, restoring all control logic, including wind-induced torque compensation, dynamic inertia update, and closed-loop rotational speed regulation.

[0091] The following is an example implementation of the "Adaptive Control Method for Luffing and Slewing Linkage of a Tower Crane for High Wind Load Environments" based on the present invention. Consider a tower crane operating at a construction site in a sea breeze environment. The crane is 80 meters tall, with a boom length of 60 meters. The current luffing angle is 40 degrees, the slewing angle is 120 degrees, the load mass is 3.2 tons, and the real-time average wind speed is 18 meters per second. The crane is equipped with ultrasonic wind measuring devices at three heights: at the tower base (0 meters), the tower cap (80 meters), and near the boom tip (approximately 78 meters high). First, the central processing unit receives wind speed data from the three wind measuring devices during the current control cycle. After timestamp alignment and clock synchronization, the data is bandpass filtered (with a lower limit of 0.1 Hz and an upper limit of 10 Hz) to remove low-frequency drift and high-frequency electrical noise. The resulting effective wind speed data are 16.5, 18.2, and 18.0 meters per second, respectively. The sliding time window width is set to 300 milliseconds to smooth the gust components. After eliminating data jump samples based on the preset standard deviation threshold, the current effective average wind speed is calculated to be 17.6 meters per second, and the wind direction is 5 degrees north-east.

[0092] Based on the current boom attitude and coordinate transformation algorithm, the effective wind speed is decomposed into a tangential component aligned with the boom's longitudinal direction of 12.3 meters per second and a normal component perpendicular to the boom plane of 9.8 meters per second. The central processing unit then searches the attitude-area database and determines the windward projected area corresponding to a 40-degree luffing angle to be 38.6 square meters. Using an aerodynamic model, the total wind load acting on the boom due to the integrated wind load field is calculated to be 2200 Newtons. Taking the horizontal projection distance of 30.2 meters from the boom's center of gravity to the center of rotation as the effective arm, the product yields a wind-induced coupling torque of 66,440 Newton meters. Furthermore, considering the load, the load mass is 3.2 tons, and its longitudinal distance is 80% of the boom's effective length, or 48 meters. Combining the current slewing angle and luffing angle, the vertical projection distance of the load's center of mass in the slewing axis coordinate system is converted to 42.5 meters, resulting in a total dynamic inertia contribution of approximately 5780 kilograms per square meter. The static moment of inertia of the tower and slewing platform is 3850 kg·m², and the boom weighs 4.8 tons. The change in inertia due to the luffing angle is calculated to be approximately 1630 kg·m². Therefore, the real-time equivalent moment of inertia for the current cycle is 3850 + 1630 + 5780 = 11260 kg·m².

[0093] On this basis, the central processing unit, in accordance with the upper-level task requirements, needs to adjust the amplitude angle from the current 40 degrees to 42 degrees in the next cycle, with the target amplitude angle acceleration planned to be 0.15 radians per square second. Multiplying this acceleration by the equivalent moment of inertia yields an inertia feedforward component of 1689 N·m. At this time, the measured speed of the amplitude motor is 0.18 radians per second, and after low-pass filtering, the damping component is 56 N·m. Based on the position and mass of the load, the gravity component in the current posture contributes a disturbance torque of 1524 N·m in the amplitude direction, and the projection of the wind-induced coupling torque in the amplitude direction is 448 N·m, resulting in a combined disturbance compensation of 1972 N·m. The error term, rate of change, and integral value of the closed-loop error signal group are synthesized into corresponding feedback terms to obtain an error correction of 432 N·m.

[0094] After combining all the above components, the original variable amplitude torque command is 1689 + 56 + 1972 + 432 = 4149 N·m. Considering the current system safety limit of 4000 N·m, the central processing unit triggers the linear peak clipping mechanism, smoothly adjusting the excess to the target limit and reporting the potential overload risk event to the system monitoring module. After soft limiting, the final variable amplitude torque command output is 4000 N·m.

[0095] The slewing control process then begins. The central processing unit schedules the wind-induced coupling torque projection in the slewing direction to be 61,000 N·m, the load weight overturning component to be 28,700 N·m, and the resulting torque to be 89,700 N·m. Dividing this value by the real-time inertia of 11,260 kg·m² yields a target angular acceleration of 7.96 radians per second squared. The slewing servo feedback indicates a current angular acceleration of 6.72 radians per second squared, with a velocity deviation of 0.34 radians per second. Combined with the historical integrated velocity of 0.23 radians per second, this forms a closed-loop error signal set. Since the current amplitude angular velocity is 0.12 radians per second, the system calculates the amplitude-slewing coupling kinetic energy term to be 0.89 N·m, corresponding to a velocity compensation of 0.08 radians per second. The resulting angular velocity correction is a combination of the original angular velocity target, the error response, and the kinetic energy compensation, resulting in a target angular velocity setpoint of 1.12 radians per second.

[0096] When this angular velocity setpoint was written into the slew servo command in the next cycle, the central processing unit detected that the angular acceleration had reached the critical threshold of 9.7 radians per square second due to a short-term peak wind speed surge of 24 meters per second. This exceeded the structural safety limit of 8.5 radians per square second, immediately triggering the extreme wind load safety backoff mechanism. The current slew angular velocity setpoint was frozen, and the luffing system decelerated to 0.05 radians per second. The hook damping valve was simultaneously instructed to increase the damping ratio, adjusting the tower top wind damper opening from 60% to 90%. When the wind speed dropped back to 15 meters per second and the load sway angle fell below 1.5 degrees, the system gradually released the backoff mechanism and resumed the closed-loop control process.

[0097] Figure 2 The results of the linkage control performance test are shown in the figure. In the luffing angle response section, the horizontal axis represents time (in seconds) with a time range of 0-10 seconds, and the vertical axis represents the luffing angle (in degrees) with a range of 0-40 degrees. The target value setting line, represented by a dashed line, shows a linear upward trend from the lower left to the upper right, starting at approximately 15 degrees and ending at approximately 35 degrees. The slope remains constant, indicating that the system's preset luffing angle gradually increases over time at a fixed rate, demonstrating the luffing mechanism's uniform motion control strategy. The actual value response curve, represented by a solid line, shows that in the initial stage (0-2 seconds), the actual angle value is significantly lower than the target setting value, resulting in a large tracking error of approximately 8-10 degrees. This phenomenon is primarily due to the inertia of the tower crane's luffing mechanism and the dynamic response delay during the system startup process. As time passes (2-6 seconds), the slope of the actual value curve gradually increases, indicating that the control system continuously adjusts control parameters through the adaptive algorithm, accelerating angle tracking. After about 6 seconds, the actual value curve basically coincided with the target value curve, and the tracking error was reduced to within 2 degrees, demonstrating excellent angle control performance and system stability.

[0098] In the slewing angular velocity response section, the horizontal axis also represents time (in seconds) with a time range of 0-10 seconds, and the vertical axis represents slewing angular velocity (in degrees / second) with a range of 0-4 degrees / second. The figure specifically highlights the wind disturbance impact zone, located between 4 and 6 seconds on the time axis. This zone is highlighted with a dashed rectangle and clearly labeled "Wind Disturbance" and "Affected Region," indicating that the system is significantly affected by external wind loads during this period. Within this impact zone, the target value is set as a horizontal constant line of 2 degrees / second, indicating that the desired slewing angular velocity should remain stable. The actual value response curve maintains a relatively stable state outside the wind disturbance zones (0-4 seconds and 6-10 seconds), with the angular velocity value remaining stable around 2 degrees / second and fluctuating by less than 0.2 degrees / second. However, within the wind disturbance influence zone (4-6 seconds), the actual value curve shows significant fluctuations, first rapidly rising from 2 degrees / second to approximately 3.2 degrees / second, then sharply decreasing to approximately 1.5 degrees / second, and finally readjusting to approximately 2 degrees / second at the end of the zone. This response characteristic has a peak deviation of 1.2 degrees / second, fully reflecting the significant impact of wind disturbances on rotational motion. It also verifies that the adaptive control algorithm proposed in this invention can effectively suppress the influence of disturbances within approximately 2 seconds, ultimately achieving stable tracking of the target angular velocity through real-time adjustment of control parameters and linkage compensation mechanisms. Figure 3This figure is a comparison of the effects of variable-leverage torque control. This figure, through the form of time-torque curves, intuitively demonstrates the significant difference in variable-leverage torque control performance between conventional control methods and the adaptive control method for tower crane variable-leverage and slewing linkage for high wind load environments described in the present invention. The horizontal axis in the figure represents the time axis in seconds, and the vertical axis represents the magnitude of the variable-leverage torque. The dashed curve represents the change in variable-leverage torque using the conventional control method, while the solid curve represents the change in variable-leverage torque using the control method of the present invention. Within the high wind load disturbance range, the torque curve of the conventional control method exhibits significant fluctuation characteristics, with a large difference between the peak and valley torque values. This indicates that the conventional method has difficulty maintaining stable torque output in the face of wind load disturbances, and is prone to large control deviations and system oscillations. In contrast, under the same high wind load disturbance conditions, the variable-leverage torque curve of the control method of the present invention remains relatively stable, and the torque fluctuation amplitude is significantly reduced. This improvement is primarily attributed to the real-time calculation and compensation mechanism of the wind-induced coupling torque and the dynamic adjustment function of the real-time equivalent moment of inertia in the present invention. By using the wind-induced coupling torque obtained in step 1 and the real-time equivalent moment of inertia determined in step 2, the system can actively compensate for wind load disturbances within each control cycle, thereby effectively suppressing torque fluctuations. Test data shows that the control method of the present invention reduces amplitude torque fluctuations by 65%, improves control accuracy by 40%, and shortens response time by 30%. These improvements in performance indicators directly reflect the control advantages of the present invention in high wind load environments and provide a strong guarantee for the safe and stable operation of tower cranes in adverse weather conditions.

[0099] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A tower crane luffing-slewing linkage adaptive control method for high wind load environments, characterized by: The method comprises: Step 1: Obtain the measured average wind speed at the tower crane's height. Combined with the crane's slewing angular velocity, boom length, luffing angle, and slewing angle, calculate the boom's windward projected area and relative wind speed to determine the wind-induced coupling torque acting on the crane's slewing axis. Step 2: Calculate the vertical projection distance from the tower crane's slewing center to the load based on the crane's inherent moment of inertia, boom deadweight, load mass, boom length, luffing angle, and slewing angle. Calculate the real-time equivalent moment of inertia generated by the boom deadweight, load mass, and their relative positions. Step 3: Obtain the original luffing torque command for the tower crane luffing motor output by multiplying the real-time equivalent moment of inertia and the luffing angular acceleration command, the viscous damping effect of the luffing mechanism, the gravitational torque generated by the load, and the projection of the wind-induced coupling torque in the luffing direction; Step 4: Using the coupled moment of inertia of the slewing axis, the calculated projection of the wind-induced coupling torque in the slewing direction, the tilting component of the gravity generated by the hoisted mass on the slewing axis, and the kinetic energy coupling effect caused by the linkage between the luffing angular velocity and the slewing angular velocity are converted into a correction value for the slewing angular velocity. This is used to update the slewing angular velocity setpoint of the tower crane in the next control cycle in real time, completing the linked adaptive control of the tower crane's luffing and slewing. The central processing unit performs the following processes: It applies multi-stage filtering to the raw wind farm data set, specifically using bandpass filtering to suppress low-frequency drift and high-frequency noise, using a sliding time window to smooth gust fluctuations, and removing instantaneous distortion samples based on a preset anomaly detection threshold to obtain effective wind speed and direction with stable statistical characteristics; According to the real-time luffing angle, slewing angle and boom length of the tower crane, the coordinate transformation algorithm is called to decompose the effective wind speed into a tangential component consistent with the longitudinal direction of the boom and a normal component perpendicular to the boom plane; based on the windward projection area database of the boom, the equivalent wind-exposed areas under different postures are matched by table lookup, and the tangential component and the normal component are combined with the equivalent wind-exposed area to construct a comprehensive wind load field reflecting the instantaneous aerodynamic state of the tower crane; the comprehensive wind load field is mapped to the slewing axis coordinate system, and according to the length of the effective arm from the boom to the slewing center and the position of the boom's center of mass, the wind-induced coupling torque acting on the slewing axis is calculated by the integration method; the tower crane's slewing angular velocity and the boom tip tangential linear velocity information are read synchronously, and the wind-induced coupling torque is coupled and corrected in combination with the real-time wind direction deviation angle to reflect the real-time influence of the slewing motion on the relative wind speed; Step 2 specifically includes: reading the inherent moment of inertia of the tower body and slewing platform, the deadweight of the boom, the length of the boom, and the geometric position of the center of mass of the boom and the center of rotation as a set of static design parameters, and locking the static design parameter set once based on the factory calibration report; obtaining the real-time value of the load mass and the instantaneous position of the load in the longitudinal direction of the boom, combining the amplitude angle and rotation angle information shared in step 1, performing coordinate transformation on the vertical projection distance of the load in the rotation plane and the action arm from the boom to the center of rotation, and constructing a dynamic position model of the load mass in the rotation axis coordinate system; according to the dynamic equivalence principle of the tower crane, the inherent moment of inertia of the tower body and slewing platform, the deadweight of the boom and its inertia component caused by the change of the amplitude angle, and the inertia contribution of the load mass to the rotation axis at the real-time position are accumulated separately, and the coupling inertia cross term between the boom and the center of mass of the load relative to the center of rotation is introduced in the accumulation process to obtain the real-time equivalent moment of inertia; Step 3 specifically includes: the central processing unit simultaneously calls the wind-induced coupling torque determined in step 1 and the real-time equivalent moment of inertia determined in step 2 in each control cycle, and calculates the original amplitude torque command output by the tower crane amplitude motor based on the established tower crane amplitude dynamics model to actively compensate for high wind load disturbances; the calculation process includes: forming a closed-loop error signal group containing error, error change rate and historical integral according to the target amplitude angle trajectory and the real-time measured amplitude angle; generating an inertia feedforward component with the real-time equivalent moment of inertia and the expected amplitude angle acceleration to offset the uncertainty of mass distribution changes; combining the damping component obtained by filtering the amplitude motor speed, the disturbance compensation amount formed by the projection of the load gravity and the wind-induced coupling torque in the amplitude plane, and hierarchically synthesizing them with the closed-loop error signal group; outputting the original amplitude torque command to complete the adaptive control of the amplitude mechanism.

2. The tower crane luffing-slewing linkage adaptive control method for high wind load environments according to claim 1 is characterized in that: Step 1: Ultrasonic wind measuring devices are deployed near the tip of the tower crane boom, at the tower cap, and at predetermined heights above the tower base to synchronously collect incoming wind speed, wind direction, and turbulent pulsation intensity as raw meteorological data. The raw meteorological data is timestamped according to a unified clock and sent to the central processing unit to form a wind field raw data set covering all measuring points within each control cycle.

3. The tower crane luffing-slewing linkage adaptive control method for high wind load environments according to claim 1 is characterized in that: The real-time value of the load mass is obtained through the load force measuring pin or wire rope tension sensor, and the instantaneous position of the load is obtained through the boom trolley stroke encoder. The central processing unit uses the pulse feedback signal of the tower crane's main hoisting drive and the low-frequency swing data collected by the boom tip inertial measurement unit to verify the tiny swing angles of the load mass in the longitudinal and lateral directions, and use this to correct the instantaneous spatial coordinates of the load center of mass. During the inertia accumulation process, the load mass position and boom boom angle inputs are subjected to exponentially weighted moving average filtering, and abnormal sampling points exceeding the range of three times the median absolute deviation are eliminated through the chi-square test to improve the statistical robustness of the inertia estimation.

4. The tower crane luffing-slewing linkage adaptive control method for high wind load environments according to claim 3 is characterized in that: At the end of each control cycle, the central processing unit performs a least squares self-calibration operation by comparing the measured angular acceleration output by the rotational angular velocity sensor with the theoretical angular acceleration predicted based on the real-time inertia model, and makes slight adjustments to the residuals of the boom center of mass position and the load center of mass position in the coordinate transformation matrix; when the numerical change of the real-time equivalent moment of inertia relative to the previous cycle exceeds the safety threshold, the inertia freezing logic is triggered and an inertia abnormality alarm is issued to the tower crane main controller.

5. The tower crane luffing-slewing linkage adaptive control method for high wind load environments according to claim 4 is characterized in that: Integral anti-saturation logic is embedded in the original variable amplitude torque command generation stage, and the integral accumulation speed is limited according to preset conditions to prevent instantaneous torque overshoot caused by strong gusts. When the original variable amplitude torque command is lower than the safety lower limit, the original variable amplitude torque command is set to zero to avoid motor jitter in the low load area. When the original variable amplitude torque command is higher than the safety upper limit, the peak is clipped according to the linear decreasing curve and a potential overload risk signal is sent. The original variable amplitude torque command processed by soft limiting is cross-checked with the DC bus voltage, brake resistor temperature rise and servo drive phase margin operating parameters. If the brake unit is found to be saturated or the bus voltage exceeds the limit, the energy dissipation priority strategy is triggered, and part of the potential energy feedback sequence is allocated to the rotation or lifting drive to achieve cross-mechanism energy balance.

6. The tower crane luffing-slewing linkage adaptive control method for high wind load environments according to claim 5, characterized in that: In step 4, the central processing unit uses the wind-induced coupling torque obtained in step 1, the real-time equivalent moment of inertia determined in step 2, and the original amplitude-changing torque command output in step 3 as input boundary conditions in each control cycle, and performs the following operations according to the tower crane rotation dynamics and energy conservation principle to achieve rotation stability control under high wind load environment: dispatch the projection data of the wind-induced coupling torque in the rotation direction, and perform vector-level synthesis with the overturning component converted from the load gravity in the same cycle to obtain the instantaneous synthetic torque describing the real-time external disturbance state; call the real-time equivalent moment of inertia. The instantaneous synthetic torque is converted into the target angular acceleration, and combined with the actual angular velocity and angular acceleration fed back by the slewing drive to form a closed-loop error signal set including the velocity deviation, acceleration deviation and historical velocity integral. According to the energy shaping idea, the coupling kinetic energy term between the luffing angular velocity and the slewing angular velocity is calculated, mapped into the angular velocity adjustment amount and superimposed on the velocity deviation to form a preliminary corrected angular velocity setting value, so as to update the slewing angular velocity setting value of the tower crane in the next control cycle in real time, and complete the linked adaptive control of the luffing-slewing of the tower crane.

7. The tower crane luffing-slewing linkage adaptive control method for high wind load environments according to claim 6, characterized in that: The central processing unit is also equipped with an extreme wind load safety de-control mechanism, which specifically includes: when the transient impact value detected by the slewing acceleration sensor exceeds the structural limit threshold, it immediately enters a three-level de-control mode, in which the first stage freezes the slewing servo angular velocity command and broadcasts a coordinated deceleration request to the variable amplitude drive module; the second stage reduces the impact of the wind-induced sudden rise on the slewing system by dynamically adjusting the horizontal damping of the hook and the opening of the tower top wind baffle; the third stage gradually releases the freeze and resumes the closed-loop adaptive correction process after confirming that the wind speed and the swing amplitude of the load have fallen back to a controllable range; during the de-control period, the residual wind-induced coupling torque in the slewing direction is continuously calculated in a low-frequency scanning manner, and normal control is automatically re-enabled when the safety window appears.

Images