A water type steel stone cage net box hoisting positioning control system

Abstract

The application relates to the technical field of water hoisting construction, and discloses a water type steel gabion net cage hoisting positioning control system which comprises a state sensing module, a dynamic anti-swing module, a flow field anti-interference module and an execution output module. The system acquires hoisting environment data and hoist posture data in real time; fluid added mass is calculated in combination with the porosity parameter of the net cage, a linear parameter variation state space model is constructed, a swing prevention instruction is solved and generated; when the net cage approaches a positioned net cage, a narrow-slit local adsorption force is solved by using an inverse dynamics disturbance observer, maximum lateral hydrodynamic lift is evaluated, and a yaw instruction and a translation instruction are dynamically distributed to proactively offset local flow field disturbance; finally, the execution output module fuses the instructions and outputs the instructions to corresponding actuators of hoisting equipment. The application overcomes the gain mismatch and sensor signal distortion problems under a complex flow field, and improves the anti-interference ability and positioning accuracy of the net cage during hoisting in a narrow-slit water area.

Description

Technical Field

[0001] This invention relates to the field of waterborne hoisting construction technology, specifically to a waterborne steel gabion hoisting and positioning control system. Background Technology

[0002] In water conservancy and water-related engineering construction, the hoisting and positioning of steel gabion cages on water is a critical operation. Due to the complex nonlinear flow field in the underwater environment, flexible hoisting systems are susceptible to hydrodynamic disturbances, resulting in low-frequency oscillations and position drift. Existing hoisting and positioning control systems typically use dynamic models with fixed parameters, failing to fully consider the time-varying characteristics of the fluid-added mass generated by the gabion cage as a permeable component under different water depths and flow velocities. This leads to control gain mismatch in the system when the sling length changes in real time or when there are sudden fluctuations in the flow field, making it difficult to maintain a stable anti-sway effect.

[0003] Meanwhile, during the lowering of the net cage to the assembly position, when the current net cage approaches the already positioned net cage, the water flow is compressed, generating a local suction force in the narrow slit. Conventional underwater velocity sensors are prone to signal distortion within the boundary layer of narrow waterways, making it impossible to accurately acquire local flow field disturbance data. This results in the current system being unable to accurately measure and compensate for this suction force. To address positional deviations, existing technologies often employ the main positioning mechanism of the lifting vessel for direct horizontal displacement correction. This macroscopic mechanical translation not only struggles to respond quickly to hydrodynamic changes but also easily induces secondary mechanical vibrations in the flexible lifting system, further reducing the spatial positioning accuracy of the net cage.

[0004] Furthermore, in large-scale engineering projects involving multiple vessels working together, the lowering of the net cages by the upstream lifting vessel causes severe water disturbance and creates wake vortices. Existing lifting control systems typically only provide independent closed-loop control for a single vessel, lacking a mechanism for data exchange and joint calculation of spatial operation status between adjacent lifting vessels. This results in the downstream lifting vessel being unable to predict and avoid the flow field distortion caused by the spread of upstream wake vortices during operation. The downstream net cages often deviate from their trajectory due to the impact of strong turbulence, severely affecting the positioning accuracy and operational safety of multi-vessel collaborative construction. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a waterborne steel gabion hoisting and positioning control system. This system solves the problems caused by complex nonlinear water flow environments, time-varying added mass of permeable gabions, and interference from local adsorption forces in narrow slits and the coupling interference of multiple ship wakes. These issues make it easy for waterborne steel gabion cages to experience low-frequency oscillations and positional drift during flexible hoisting, making it difficult to ensure high-precision spatial positioning and safe collaborative operations.

[0006] To achieve the above objectives, the present invention provides a waterborne steel gabion hoisting and positioning control system, comprising a status sensing module, a dynamic anti-sway module, a flow field anti-disturbance module, and an execution output module.

[0007] The state awareness module acquires real-time data on water flow, net cage depth, sling length, net cage coordinates, and lifting equipment attitude from the lifting environment. Specifically, it obtains vertical water flow velocity profile data (as the water flow data) through underwater acoustic current measurement equipment, acquires sling length data through the encoder of the hoisting winch, obtains the ship's reference coordinates and calculates the net cage coordinates through a global navigation satellite system receiver, and acquires net cage depth and attitude data through a depth gauge and inertial measurement unit mounted on the lifting equipment. Attitude data includes pitch angle, roll angle, angular velocity vector, and three-axis acceleration vector. This module establishes a multi-source spatiotemporal aligned data foundation for the system.

[0008] The dynamic anti-sway module calculates dynamic damping parameters based on sling length and water flow data, and outputs anti-sway commands based on these parameters. Specifically, the module extracts corresponding aligned equivalent flow velocity data from the water flow data based on the gabion depth data, and calculates the fluid-added mass by combining this with the pre-set porosity parameters of the gabion. The system updates the fluid-added mass as a mass parameter into the system's pre-set mass matrix, and simultaneously constructs a linear parameter variation state-space model using the real-time updated sling length data as a scheduling variable. Based on this model and the real-time angular velocity vector extracted from the attitude data, the module solves for the regulator gain matrix and generates anti-sway commands. This calculation process quantifies the dynamic inertia characteristics of the permeable gabion in a complex flow field, overcoming the gain mismatch problem of the fixed parameter model under conditions of rope length and abrupt flow field changes.

[0009] The flow field disturbance mitigation module compares the coordinate data of the gabion cages with a preset 3D construction grid coordinate library. When it determines that the distance between the current gabion cage and the already positioned gabion cages is less than a set threshold, it assesses the local flow field disturbance force on the gabion cage based on water flow data and attitude data. This local flow field disturbance force manifests as the local suction force in the slit caused by adjacent gabion cages. Since conventional flow velocity sensors are prone to signal distortion within the slit, the flow field disturbance mitigation module incorporates an inverse dynamics disturbance observer. Utilizing the components of the triaxial acceleration vector in the horizontal dimension of the attitude data, the preset total system mass, and the acquired control driving force, it calculates the local suction force in the slit, representing the real-world environmental disturbance, in real time through inverse dynamics equations.

[0010] After acquiring the local suction force of the slit, the flow field disturbance suppression module performs dynamic allocation logic calculation to offset the hydrodynamic compensation required to counteract the disturbance force: the system calculates the maximum lateral hydrodynamic lift that the cage can generate based on the aligned equivalent velocity data extracted from the water flow data and the maximum yaw angle allowed by the attitude adjustment mechanism. When the maximum lateral hydrodynamic lift is greater than or equal to the local suction force of the slit, the system only generates an attitude yaw command, using the lateral hydrodynamic lift generated by the cage tilting towards the water to offset the local suction force of the slit. At this time, the flow field disturbance suppression module, based on the fluid lift inverse solution model, uses the aligned equivalent velocity data, the pre-calibrated attitude angle and lift coefficient mapping relationship, and the local suction force of the slit to be offset to inversely calculate the target compensation attitude angle as the attitude yaw command output. When the maximum lateral hydrodynamic lift is less than the local suction force of the slit, it indicates that attitude yaw alone cannot completely resist environmental disturbances. The system locks the attitude yaw command to the control signal corresponding to the maximum yaw angle and converts the difference between the local suction force of the slit and the maximum lateral hydrodynamic lift into a translation command and outputs it.

[0011] To address the secondary mechanical oscillations caused by translation commands, the flow field disturbance suppression module incorporates a zero-vibration input shaper. When the system outputs a translation command, the zero-vibration input shaper calculates the real-time natural frequency based on the sling length data and decomposes the step-type translation command into multiple pulse displacement sequences with preset time delays. The time delay of the pulses is used to cancel out the primary and secondary oscillations caused by the translation of the main positioning mechanism.

[0012] When the distance between the current net cage and the already positioned net cage is determined to be greater than or equal to a set threshold, it indicates that the net cage is in the free flow field region. The flow field disturbance rejection module stops outputting attitude yaw and translation commands. The system extracts the corresponding aligned equivalent flow velocity data from the water flow data based on the net cage depth data, and calculates the macroscopic water flow thrust by combining the net cage's projected area and porosity parameters, generating a feedforward displacement command.

[0013] The execution output module fuses the anti-sway command with the displacement command used to control the horizontal movement of the lifting equipment, and outputs it to the main positioning mechanism of the lifting equipment. The displacement command includes translational command or feedforward displacement command output by the flow field anti-disturbance module. Simultaneously, the execution output module outputs the attitude yaw command to the attitude adjustment mechanism of the lifting device. At the underlying equipment mapping level, the main positioning mechanism is the slewing mechanism and luffing mechanism of the lifting vessel, and the attitude adjustment mechanism is a hydraulic lifting frame equipped with four independent hydraulic leveling cylinders at the corners. The execution output module converts the fused command into pulse-width modulated electrical signals and sends them to the slewing mechanism and luffing mechanism, and converts the attitude yaw command into control electrical signals and sends them to the four independent hydraulic leveling cylinders at the corners, achieving isolated and synchronous execution of macroscopic position compensation and microscopic attitude adjustment.

[0014] Furthermore, for multi-vessel collaborative operations, the system includes a network scheduling module. This module connects to the state perception module and receives the operating coordinates and lowering depths of adjacent lifting vessels within the same construction area. Based on the cage coordinate data and operating coordinates, the system identifies the upstream and downstream lifting vessels within the same construction area. When the downstream lifting vessel is within the diffusion envelope of the wake vortex generated by the upstream lifting vessel lowering the cage, it issues a depth hovering command to the winch control system of the downstream lifting vessel, continuing until the local flow field is determined to have returned to steady state before releasing the hovering. This logic, through real-time calculation of the spatial state of multiple vessels, avoids the strong turbulence interference caused by wake vortices in terms of timing.

[0015] This invention provides a lifting and positioning control system for steel gabion cages on water. It has the following beneficial effects: 1. This invention extracts aligned equivalent flow velocity data based on the depth data of the permeable gabion and calculates the fluid-added mass by combining it with porosity parameters. Then, it uses this data, together with real-time sling length data, as variables to construct a linear parameter variation state space model to solve the regulator gain matrix and generate anti-sway commands. This accurately quantifies the dynamic inertia of the permeable gabion at different water depths, overcomes the gain mismatch problem of conventional fixed parameter control models under rope length and flow field abrupt changes, and effectively improves the suppression accuracy of flexible swaying of the gabion in complex flow field environments.

[0016] 2. This invention uses a built-in inverse dynamics interference observer to inversely calculate the local adsorption force of the slit based on the system's preset total mass, control driving force, and acceleration vector in the attitude data. It also evaluates the maximum lateral hydrodynamic lift based on the equivalent flow velocity data to dynamically allocate attitude yaw and translation commands. This allows the system to avoid the signal distortion defects of conventional flow measurement equipment in slit waters and directly utilize the lateral lift generated by the tilting of the net cage to actively counteract local flow field disturbances, thereby improving the spatial positioning anti-interference capability of the components when they approach the already positioned net cage.

[0017] 3. This invention receives the working coordinates and lowering depth of adjacent lifting vessels in the construction water area in real time through a network scheduling module. When it is identified that the downstream lifting vessel is within the diffusion envelope of the wake vortex generated by the upstream lifting vessel lowering the net box, a depth hovering command is issued to the downstream vessel. In this way, by jointly solving the spatial state of multiple vessels, the upstream flow field distortion region is actively avoided in time sequence, eliminating the negative interference of strong wake turbulence on the downstream lifting positioning accuracy, and ensuring the safety and stability of multi-vessel collaborative construction operations. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall architecture and control flow of an embodiment of the present invention; Figure 2 This is a schematic diagram of the hardware layout and data flow of the state perception module in an embodiment of the present invention; Figure 3 This is a control logic architecture diagram of the dynamic anti-sway module according to an embodiment of the present invention; Figure 4 This is a logic diagram of the underlying boundary avoidance and dynamic control allocation strategy of the flow field anti-disturbance module in an embodiment of the present invention; Figure 5 This is a logic diagram of the free flow field feedforward compensation for the flow field disturbance rejection module in an embodiment of the present invention; Figure 6 This is a logic diagram of multi-ship coordination and wake vortex anti-disturbance scheduling in an embodiment of the present invention; Figure 7 This is a logic diagram of system collaborative execution and underlying hardware driver allocation in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] See attached document Figure 1 In this embodiment, the physical hardware architecture of a floating steel gabion hoisting and positioning control system relies on a floating hoisting vessel and its supporting lifting equipment system. The underlying actuators of the system mainly consist of a main positioning mechanism and an attitude adjustment mechanism.

[0021] Specifically, the main positioning mechanism is the slewing mechanism and luffing mechanism of the crane ship. It contains a slewing motor and a luffing hydraulic cylinder, which are mainly used to receive control commands from the upper level and drive the lifting point to move in a macroscopic position in the horizontal plane.

[0022] The attitude adjustment mechanism uses a hydraulic lifting frame equipped with four independent hydraulic leveling cylinders at the corners. The lower end of the lifting frame is used to form a rigid connection with the gabion box. The system adjusts the tilt angle and yaw attitude of the underwater gabion box by independently controlling the extension and retraction stroke of the four leveling cylinders.

[0023] To address the engineering requirements of handling variable underwater flow fields, this system incorporates multiple logic control modules, including a state sensing module, a dynamic anti-sway module, a flow field disturbance rejection module, an execution output module, and a network scheduling module. These modules are interconnected via industrial network nodes according to a bus protocol, employing a combination of closed-loop feedback and flow field feedforward to achieve hoisting and positioning.

[0024] As a preferred approach, the overall system workflow consists of the following steps: S100. In order to obtain the external flow field environment and internal mechanical operation boundary conditions required for the operation of the control system, the state perception module continuously performs multi-source data acquisition.

[0025] This module uses underwater acoustic flow measurement equipment and satellite navigation sensors to acquire real-time data on water flow, net cage depth, sling length, net cage coordinates, and lifting gear attitude.

[0026] Given that the system involves multiple heterogeneous sensors, such as acoustic flow measurement equipment and inertial measurement units, their sampling frequencies inherently differ. To avoid errors in subsequent matrix operations caused by phase lag, the state perception module interpolates and aligns the timestamps of each sensor based on the global clock before data fusion, ensuring strict correspondence between the state parameters of each dimension in terms of time axis and spatial conditions.

[0027] S200. Since the swing frequency and amplitude characteristics of the flexible sling in the underwater environment are highly dependent on the current real-time rope length and medium damping, the dynamic anti-sway module intervenes in real time based on the various alignment data obtained by the state perception module.

[0028] This module calculates dynamic damping parameters based on sling length data and water flow data, and outputs anti-sway commands within the state space framework to consume the flexible sway energy of the hoisting system.

[0029] S300, the components already in place at the bottom of the engineering water area will change the local flow field around them. This change will create a sudden change in flow velocity and pressure difference adsorption effect in the slits.

[0030] Based on this phenomenon, the system needs to determine the current working range of the cage. The flow field disturbance rejection module compares the cage coordinate data with the system's preset three-dimensional construction grid coordinate library, calculates the spatial distance between the current cage and the nearest cage that has been placed, and determines whether the distance is less than a set threshold.

[0031] The aforementioned threshold is not a fixed constant. It is calculated by combining the wake diffusion theory in fluid mechanics with the equivalent width of the water-facing side of the actual cage size. It is used to define the spatial boundary where local flow field distortion can cause substantial interference to the cage.

[0032] Meanwhile, to avoid misjudgments caused by single-point coordinate jumps, the spacing determination logic introduces a time window moving average mechanism to verify multi-dimensional parameters, rather than relying solely on a single instantaneous distance extreme value.

[0033] Based on the judgment results of S300, the flow field disturbance rejection module executes a differentiated control strategy for different local flow field characteristics.

[0034] S410. When it is determined that the distance between the current cage and the cage that has been placed is greater than or equal to the set threshold, it indicates that the cage is in the free flow field region and is not disturbed by the boundary flow field of the existing components.

[0035] At this point, the system extracts the corresponding real-time flow velocity based on the depth of the net cage, calculates the macroscopic water flow thrust based on the net cage's upstream area, and directly generates a feedforward displacement command. This compensation action is used to offset the steady resistance exerted by the steady-state water flow on the lifting device system in advance by utilizing the macroscopic translation of the main positioning mechanism.

[0036] S420. When it is determined that the distance between the current cage and the cage that has been placed is less than the set threshold, it indicates that the cage has entered the slit interference zone, and conventional macroscopic feedforward calculation is no longer applicable.

[0037] The flow field disturbance mitigation module then assesses the local flow field disturbance force experienced by the cage based on flow and attitude data. Based on this local flow field disturbance force, the system calculates the hydrodynamic compensation required to counteract its effects, and then converts this compensation into attitude yaw commands required for the hydraulic lifting frame's movement, or translation commands required for the crane vessel's movement, at the execution end.

[0038] S500: To ensure that each control logic is smoothly implemented at the layer, the execution output module schedules and processes the above heterogeneous instructions.

[0039] This module fuses the anti-sway commands output by the dynamic anti-sway module with displacement commands used to control horizontal movement. These displacement commands include feedforward displacement commands or translation commands depending on the current operating condition. The essence of the fusion process is that the anti-sway commands provide damping compensation for high-frequency transient oscillations, while the displacement commands provide position servoing for low-frequency steady-state drift. The two are superimposed in linear space without interfering with each other.

[0040] The fused command is output to the main positioning mechanism to perform macroscopic compensation; At the same time, the attitude yaw command is output to the attitude adjustment mechanism to execute the micro attitude deflection of the cage.

[0041] The motion compensation of the main positioning mechanism and the yaw compensation of the attitude adjustment mechanism work together to achieve high-precision positioning of the cage during lowering.

[0042] In addition to executing the aforementioned individual control procedures, in scenarios involving multiple vessels operating in a continuous area, the wake vortices generated when the upstream crane vessel lowers the net cages can severely impact the downstream waters. The network scheduling module receives the operational coordinates and lowering depths of adjacent crane vessels within the same construction area in parallel.

[0043] When the downstream crane vessel is located within the diffusion envelope of the wake vortex of the upstream crane vessel based on the coordinate data of the cage, the module directly sends a deep hovering command to the hoisting control system of the downstream crane vessel until the system determines that the local flow field has returned to steady state and then releases the hovering lock, thereby avoiding unpredictable strong turbulence interference in advance at the time logic level.

[0044] See attached document Figure 2 In this embodiment, in order to determine the flow field environment and mechanical operation boundary in the control process, the system clarifies the data acquisition mechanism and hardware support structure of the internal data of the state perception module.

[0045] Based on the perception logic from macroscopic environment to microscopic load, the data acquisition process of the state perception module is refined into the following collaborative sub-steps.

[0046] S110. Since the force of water flow varies with water depth and exhibits a shear distribution pattern, the flow velocity at a single depth is difficult to characterize the overall hydrodynamic disturbance.

[0047] Based on the above correlations, the system needs to obtain the hydrological distribution of the entire basin. The state perception module is equipped with an underwater acoustic current measurement device installed on the bottom of the hull of the lifting vessel, which uses an acoustic Doppler current profiler.

[0048] The profiler emits acoustic pulses downwards and receives the scattered echoes from suspended particles in the water. It calculates and outputs vertical flow velocity profile data using the Doppler frequency shift principle. The system assumes a uniform horizontal flow field distribution within the local operating area, and uses this vertical profile data to directly characterize the hydrological features of the actual spatial location of the net cage.

[0049] This data forms a one-dimensional discrete array with water depth as the independent variable, which forms the data basis for subsequent calculations of fluid-added mass.

[0050] S120. In order to establish the absolute spatial mapping relationship between the lifting equipment and the operating water area, the status perception module synchronously performs the acquisition of spatial position parameters.

[0051] The system is equipped with an absolute photoelectric encoder on the drum side end face of the hoisting winch. By reading the drum rotation angle and combining it with the drum layer winding radius, the system calculates and obtains the sling length data.

[0052] Meanwhile, the system is equipped with a high-precision tilt sensor at the hinge point of the luffing hydraulic cylinder to obtain the luffing angle of the crane boom in real time.

[0053] A global satellite navigation system receiver is installed at the reference point of the lifting vessel to obtain the vessel's reference coordinates. For the three-dimensional coordinate calculation process of estimating the absolute underwater position of the net cage from the vessel's reference coordinates, crane boom geometry, slewing angle, real-time luffing angle, and sling length, those skilled in the art can programmatically solve the problem based on the fundamental rigid body kinematics forward equations.

[0054] The system outputs cage coordinate data containing longitude, latitude and elevation information in real time, which serves as the spatial basis for triggering obstacle avoidance logic.

[0055] S130. Due to water flow turbulence and component interference, the net cage will generate complex spatial oscillations. The system implements micro-state monitoring at the end of the lifting device to capture this dynamic characteristic.

[0056] As a preferred method, the hydraulic lifting frame is rigidly locked to the gabion box, and the status sensing module integrates a depth gauge and an inertial measurement unit inside the hydraulic lifting frame.

[0057] The depth gauge outputs the depth data of the net cage based on changes in external water pressure. The inertial measurement unit outputs the attitude data of the lifting device through its built-in microelectromechanical gyroscope and accelerometer.

[0058] The attitude data consists of pitch angle, roll angle, angular velocity vector, and triaxial acceleration vector, and is used to directly quantify and reflect the real-time dynamic load state of the cage under the influence of the flow field.

[0059] S140. Since acoustic devices and inertial measurement units operate on different principles, their inherent sampling periods differ.

[0060] If asynchronous data is directly introduced into the subsequent state-space model, a slight time lag will cause the discretization of matrix operations to diverge.

[0061] Based on the above engineering constraints, the state perception module embeds clock synchronization logic, and performs timestamp interpolation alignment between low-frequency water flow profile data and high-frequency attitude data based on the principle of linear interpolation.

[0062] The specific alignment calculation process is as follows: Let the timestamp of the high-frequency signal output by the inertial measurement unit be... The timestamps of two consecutive low-frequency profile signals output by the underwater acoustic current measurement equipment are respectively and And satisfy .

[0063] The specific water layer flow velocity values ​​extracted from the low-frequency signal within this time interval are as follows: and The state-aware module calculates the timestamp. Imprint the aligned equivalent flow rate data The calculation formula is as follows: ; Among them, variables The function to find the maximum value; constant This represents the smallest positive tolerance set by the system based on the hardware clock resolution. Its value is determined according to the instruction cycle of the system's microprocessor; in this embodiment, it is set to 10. -6 s.

[0064] This tolerance mechanism is used when bus data congestion causes the sensor to repeatedly output the same timestamp data, i.e. equal When forced to take boundary values ​​in the denominator This avoids system crashes caused by the denominator being equal to zero in division operations at the algorithm's underlying level.

[0065] Through this numerical interpolation operation, the system ensures that the spatial acceleration of the cage and the fluid kinetic energy parameters at the same time section are strictly at the same time section before entering the dynamic anti-sway module for joint solution.

[0066] See attached document Figure 3 In this embodiment, due to the combined interference of wind, waves, and currents when the crane vessel operates on the water surface, the long-distance suspended cable system will produce flexible low-frequency oscillations similar to a simple pendulum. Conventional constant-parameter control methods are difficult to adapt to the complex working conditions of underwater operations.

[0067] The reason is that the gabion mesh has a unique permeable structure, which makes the additional mass generated by the synchronous movement of the surrounding water when it moves underwater not a constant value.

[0068] Based on the understanding of the aforementioned engineering phenomena, the system's embedded dynamic anti-sway module relies on high- and low-frequency fused data provided by the state perception module to perform variable-parameter anti-sway intervention. The specific control process of this module is detailed as follows: S210. As a preferred method, in order to accurately quantify the true inertial characteristics of the cage when it moves underwater, the dynamic anti-sway module calculates the fluid-added mass based on fluid dynamics theory.

[0069] The system extracts the aligned equivalent velocity data of the corresponding water layer based on the currently acquired gabion depth data. Based on the velocity characteristics of this spatial profile, and combined with the preset porosity parameters of the gabion cage, the system calculates the fluid-added mass caused by the water viscosity effect. The calculation expression is as follows: ; in, Represents timestamp The added mass of the fluid is engraved; The dimensionless additional mass coefficient is pre-calibrated through a water tank test based on potential flow theory. The fluid density representing the operating water area; The volume of the outer enclosure of the gabion cage; The porosity parameter of a gabion mesh cage represents the ratio of the effective permeable space volume to the total enclosed volume. Its value is strictly defined within the open interval (0,1). Here is the hydrodynamic correction function, where The dynamic pressure reduction factor is determined based on the structural hydrodynamics, and its value range is usually set between 0.05 and 0.15 depending on the mesh size of the gabion.

[0070] This formula, by introducing a porosity parameter and a quadratic dynamic pressure variable based on real-time flow velocity, corrects the theoretical constant inertia into a time-varying parameter that fits the complex nonlinear underwater working conditions, thereby accurately describing the evolution law of the dynamic inertia of the cage under actual water disturbance.

[0071] S220. As the hoisting and lowering process is accompanied by the continuous operation of the hoisting winch, the length of the sling is in a state of continuous dynamic change. This change in macroscopic dimensions will directly affect the natural period of the pendulum system.

[0072] To establish an accurate foundation for dynamic control, the dynamic anti-sway module constructs a linear parameter variation state-space model dependent on real-time variables based on the Lagrange dynamic equations. The system adds the fixed dry mass of the cage to the time-varying fluid-added mass obtained from the above formula to form the current total mass parameter of the system, and updates the system's mass matrix as a non-constant element.

[0073] Simultaneously, the module extracts the real-time updated sling length data as an independent scheduling variable driving the system evolution, thereby establishing a model describing the dynamic characteristics of the coupled system of the lifting vessel, sling, and cage. The structural expression of its state-space equations is as follows: ; in, The state vector of the system contains the horizontal displacement and speed of the lifting equipment, as well as the spatial swing angle and angular velocity of the sling. The input constraint variable is the output of the control system, which physically represents the desired driving acceleration of the main positioning mechanism motor. Represents the length of the sling. Added mass to fluid The system state matrix evolves and updates in real time; The input control matrix depends on the scheduling variables.

[0074] Since the system state matrix is ​​constructed as a continuous function of the scheduling variables, those skilled in the art can use polyhedral vertex mapping expansion and Jacobian linearization methods to reduce the dimensionality of the local operating points of this parametric system.

[0075] This variable parameter model avoids the gain mismatch and system divergence problems that occur with a single fixed parameter model when the rope length changes or the water flow changes abruptly.

[0076] S230. After the system state space is constructed, in order to generate specific mechanical action commands to consume the flexible oscillation kinetic energy of the system, the dynamic anti-sway module performs closed-loop feedback calculation based on the above linear parameter change model.

[0077] During this process, the system performs a full-rank check on the control pair consisting of the parameter-dependent system state matrix and the input control matrix, that is, verifies whether the controllability matrix of the current local operating point is full-rank.

[0078] To ensure the robustness of the control system and address safety hazards caused by matrix singularity, the module will trigger a degradation protection mechanism when the controllability matrix is ​​not full rank or close to singular due to extremely short slings or parameter jumps.

[0079] Specifically, the system freezes the current gain calculation, directly retrieves and maintains the state feedback gain parameters from the previous effective calculation cycle, until the system state variables leave the singular operating range.

[0080] Under the premise that the system matrix is ​​non-singular and the current state is completely controllable, the module combines the real-time angular velocity vector extracted from the attitude data as an external feedback term to solve the parameter-dependent regulator gain matrix.

[0081] The specific calculations rely on a linear quadratic regulator algorithm. The system obtains the optimal state feedback gain matrix by solving the algebraic Riccati equation in real time. The dynamic anti-sway module generates anti-sway commands to suppress the flexible swaying of the hoisting equipment by performing a negative feedback product operation between the system's current state vector and the obtained regulator gain matrix.

[0082] At the execution level, the anti-sway command is represented as a compensation displacement signal that drives the slewing motor or luffing hydraulic cylinder of the crane vessel.

[0083] This command drives the main positioning point of the lifting equipment to generate a small accompanying damped motion in the horizontal plane that is in the same frequency and opposite in phase to the swing direction of the underwater net cage. The mechanical work is used to extract and offset the excitation energy injected into the flexible system by the water flow in real time, ensuring the stable convergence of the posture during the lifting process.

[0084] See attached document Figure 4 In this embodiment, when the lifting equipment lowers the gabion cage to an area close to the existing seabed structure, the hydrodynamic characteristics of the working environment will change abruptly.

[0085] To address the localized suction force in the slit and the resulting positional shift, the system's embedded flow field disturbance rejection module performs a series of closed-loop operations, from disturbance reconstruction to dynamic allocation of control commands. The specific operational logic and implementation process of this core control zone are detailed below: S310. After determining that the distance between the current cage and the cage already in place is less than the set threshold, the cage officially enters the slit interference zone.

[0086] Due to the physical environment, the flow velocity of water surges when it flows through the narrow gap between the two components. Based on Bernoulli's principle of fluid mechanics, the dynamic pressure on the surface of this side drops sharply, thereby generating a lateral adsorption effect that pulls the current cage toward the cage that is already in place.

[0087] Because the adsorption force is an unsteady external environmental disturbance, conventional flow velocity sensors are prone to signal distortion due to the complex eddies within the slit, making direct measurement difficult.

[0088] Based on this engineering measurement constraint, the flow field disturbance rejection module adopts a scheme of inverse dynamic state deduction.

[0089] The system introduces an inverse dynamics disturbance observer, using the cage itself as a large-scale environmental sensing carrier. Utilizing high-frequency triaxial acceleration from attitude data, the current total mass of the system, and known control driving forces, it reconstructs in real-time the local adsorption force of the slit, representing the actual environmental disturbance, through inverse dynamics equations. Its core observation equation is: ; in, Represents timestamp The local adsorption force of the slit, calculated by reconstruction, is recorded. The total mass of the system includes the dry mass of the cage and the real-time fluid-added mass. The components of the triaxial acceleration output by the inertial measurement unit in the horizontal transverse dimension; This is the known execution force feedback issued and transformed by the system in the previous control cycle.

[0090] To avoid abrupt changes in reconstruction force caused by a single high-frequency acceleration noise, the system performs preliminary calculations on the reconstructed force. A first-order low-pass filter is then applied to obtain a smooth and meaningful reference value for environmental disturbances.

[0091] S320. After accurately acquiring the local adsorption force of the slit, the system faces the scheduling problem of heterogeneous actuators. The attitude adjustment mechanism is responsive but lacks the ability to change its macroscopic position, while the main positioning mechanism has strong translational compensation capabilities but is prone to causing secondary flexible oscillations.

[0092] As a preferred intervention criterion, the system prioritizes using the fluid lift generated by changing the orientation of the cage to counteract the lateral adsorption force.

[0093] The flow field disturbance rejection module calculates the maximum lateral hydrodynamic lift that the cage can generate based on the extracted aligned equivalent flow velocity data and the maximum yaw angle allowed by the attitude adjustment mechanism.

[0094] Its calculations rely on the standard hydrodynamic lift model: ; in, This is the maximum lateral hydrodynamic lift. For fluid density; The maximum yaw angle is determined by the mechanical interference boundary between the hydraulic lifting frame and the universal joint. In this embodiment, based on the interference limit of the mechanical drawings, it is set within the range of 15° to 20°. The dimensionless lift coefficient corresponding to the maximum yaw angle is obtained in advance by scale-down tests in the engineering water tank; The effective lateral water-facing area of ​​the gabion mesh cage affected by its attitude deflection; To align equivalent flow velocity data.

[0095] Through this calculation, the system has determined the maximum resistance boundary that can be provided by attitude deflection alone at the current level.

[0096] S330. Based on the above boundary conditions, the system compares and splits the filtered slit local adsorption force with the maximum lateral hydrodynamic lift.

[0097] To avoid frequent control mode switching due to minor fluctuations in disturbance forces near the critical point, the flow field disturbance rejection module incorporates a hysteresis logic interval including upper and lower boundaries. When the maximum lateral hydrodynamic lift is stably greater than the local suction force of the slit and exceeds the upper boundary of the hysteresis interval, it indicates that the cage has sufficient attitude self-adjustment margin.

[0098] At this point, the system only generates attitude yaw commands. The flow field disturbance rejection module calls the fluid lift inverse model and, using the aligned equivalent velocity data and the local adsorption force of the slit to be offset, inversely calculates the target compensation attitude angle.

[0099] This reverse calculation process utilizes a pre-defined discrete data table of attitude angles and lift coefficients stored in system memory, combined with a cubic spline interpolation algorithm, to ensure the accuracy of the nonlinear mapping. The calculated target compensated attitude angle is directly output as the attitude yaw command.

[0100] If the current force comparison result is between the upper and lower limits of the hysteresis interval, the system forcibly freezes the state machine switching, maintaining the output mode of the previous control cycle unchanged, thereby filtering out high-frequency judgment oscillations caused by local eddies.

[0101] S340. When the bottom flow velocity is low or the slit is extremely narrow, causing a sharp increase in the local adsorption force of the slit, making the local adsorption force of the slit greater than the maximum lateral hydrodynamic lift and breaking through the lower boundary of the hysteresis interval, the system determines that the attitude mechanism has reached the resistance limit.

[0102] Under this condition, the flow field disturbance rejection module will forcibly lock the attitude yaw command to the control signal corresponding to the maximum yaw angle, ensuring that the system maintains the basic hydrodynamic pressure resistance state by utilizing the end stroke.

[0103] Meanwhile, the system extracts the difference between the local adsorption force of the slit and the maximum lateral hydrodynamic lift, and combines it with the translational stiffness coefficient of the working plane where the main positioning mechanism is located, and linearly converts it into a translational command for driving the crane ship to move.

[0104] S350 Once the main positioning mechanism receives the translation command and generates a horizontal displacement, this step-like macroscopic motion will inevitably inject excitation energy into the flexible sling, triggering a translational oscillation.

[0105] To avoid frequency conflicts between this command and the aforementioned dynamic anti-sway module, a zero-vibration input shaper is connected in series inside the flow field disturbance rejection module.

[0106] The shaper breaks down a single step drive signal into multiple pulses at specific time intervals, using the oscillations generated by subsequent pulses to cancel out the residual oscillations of preceding pulses.

[0107] In practice, the zero-vibration input shaper calculates the current real-time natural frequency based on the sling length data provided by the state sensing module, and decomposes the original translation command into two pulse displacement sequences.

[0108] The formula for calculating the time delay between two main pulses is: ; in, This is the pulse time delay. This is real-time data on the length of the slings; The acceleration due to gravity is constant; For water density, This represents the equivalent density of the gabion mesh. The equivalent gravity correction factor is used to account for underwater buoyancy.

[0109] Considering the crane vessel's own draft and safety operating regulations, the length of the slings... When entering underwater operation mode, the length is always greater than the set safety lower limit, so the value inside the square root is always a positive real number.

[0110] The shaping mechanism uses a second pulse signal with a delay of half a self-oscillation period to induce a secondary oscillation with the same amplitude as the first oscillation caused by the first pulse but with a strict phase difference of 180°. The residual oscillation is canceled at the execution end by relying on the principle of anti-phase superposition, thereby ensuring the smooth execution of the translational obstacle avoidance action.

[0111] See attached document Figure 5 In this embodiment, since underwater ranging sensors are easily affected by suspended objects or bubbles, generating high-frequency jump signals, the system avoids relying solely on the extreme distance values ​​at a single moment for pattern determination.

[0112] The flow field disturbance rejection module introduces time window weighting logic to continuously monitor the distance between the current cage and the cages already in place.

[0113] When the flow field anti-interference module confirms, based on the aforementioned spacing determination logic, that the spacing is stably greater than or equal to the set threshold within a preset time window, it indicates that the cage is in a free flow field region that is not interfered with by the boundary flow field of the existing components.

[0114] Since the slit adsorption effect disappears at this point, the main external disturbance experienced by the cage underwater is transformed into steady resistance exerted by the steady-state water flow.

[0115] To prevent long-term positional drift of the hoisting system caused by this steady resistance, the system triggers macroscopic free flow field feedforward compensation logic, which counteracts fluid resistance by applying a reverse bias displacement in advance. The specific execution of this compensation logic is detailed as follows: S410. As a preferred method, in order to quantify the actual thrust of the free flow field on the cage, the flow field disturbance rejection module calculates the macroscopic water flow thrust based on the fluid dynamics resistance model.

[0116] The system extracts the equivalent flow velocity data that has been obtained and time-aligned in the aforementioned steps, and calculates the positive force of the current field on the upstream surface of the gabion under the current water depth by combining the geometric parameters of the gabion.

[0117] The calculation formula is as follows: ; in, Represents timestamp Macroscopic water flow thrust; parameters For fluid density, Porosity parameter To align equivalent flow velocity data.

[0118] Newly introduced parameters This is a dimensionless fluid flow resistance coefficient. Its specific value is determined by looking up the Reynolds number of the gabion shape from a hydrodynamic empirical table. For a standard rectangular gabion, this coefficient is usually set in the range of 1.2 to 1.8 in this embodiment. It represents the orthogonal projected area of ​​the gabion cage in the direction perpendicular to the water flow.

[0119] This porosity parameter is used to eliminate the void area in the permeable gabion that does not actually block the water flow, thereby obtaining the equivalent water-facing surface resistance that conforms to the actual underwater working conditions and avoiding overcompensation due to overestimation of fluid thrust.

[0120] S420. Because the flexible suspension system consisting of the slings and the net cage exhibits spring-like restoring force characteristics in the horizontal direction, the steady-state macroscopic water flow thrust will inevitably cause the net cage to deviate from directly below the lifting vessel.

[0121] To eliminate this steady-state position deviation in advance, the system needs to convert the calculated hydrodynamic force into the displacement compensation of the lifting vessel.

[0122] Based on the above correlation, the flow field disturbance rejection module calculates the horizontal equivalent stiffness of the current system in the controlled direction according to the static equilibrium principle of a simple pendulum, and derives the feedforward displacement command accordingly.

[0123] The conversion formula is as follows: ; ; in, Represents the horizontal equivalent stiffness of the hoisting system at the current timestamp; Parameter For the total mass of the system, This is real-time data on the length of the slings.

[0124] Considering the potential for system divergence due to sensor malfunctions resulting in zero output or extremely short rope lengths, a maximum value function is introduced into the denominator of the formula. and the lower limit protection constant.

[0125] in, The minimum safety cable length is set based on the draft of the lifting vessel. The minimum allowable stiffness tolerance for the system is set to 0.01 N / m in this embodiment.

[0126] By introducing the aforementioned boundary protection, the system avoids the risk of computational crash caused by the denominator approaching zero during division operations at the algorithm's underlying level.

[0127] The calculated feedforward displacement command is represented as the horizontal distance that the main positioning point of the hoisting vessel needs to move towards the waterfront.

[0128] Through the above causal deduction, the system establishes a deterministic linear mapping relationship between the macroscopic flow field disturbance force and the spatial compensation displacement of the main positioning mechanism.

[0129] S430. After calculating the compensation displacement, the smoothness of the execution layer is equally important. If the feedforward displacement command is directly sent to the main positioning mechanism as a step signal after it is obtained, it is easy to induce the swaying of the lifting ship and cause hydraulic shock to the actuator.

[0130] For smooth transition and trajectory planning of displacement commands, those skilled in the art can use an S-shaped velocity curve planning algorithm based on maximum velocity and maximum acceleration constraints to shape the commands.

[0131] After smoothing, the feedforward displacement command is output to the execution output module, driving the slewing mechanism and luffing mechanism of the crane vessel to move a specified distance in advance in the direction of the water flow.

[0132] This compensation action utilizes the macroscopic translation of the main positioning point on the bottom of the ship to construct a preset flexible tilting posture, causing the sling, which is under its own weight tension, to generate a horizontal recovery force downstream of the water flow.

[0133] The recovery force and the macroscopic water flow thrust cancel each other out in the horizontal space, thus ensuring that the actual vertical landing point of the underwater cage is not affected by the steady water flow and is locked at the positioning coordinates of the seabed target.

[0134] See attached document Figure 6 In this embodiment, large-scale underwater foundation protection projects often face complex working conditions where multiple crane vessels operate concurrently in the same water area.

[0135] Due to the continuity and viscosity of fluids, the gabion cages lowered by the upstream crane vessel will form a wake vortex zone that spreads backward under the shearing action of the water flow.

[0136] If the cages to be lowered by downstream vessels cut into the region of flow distortion, the alternating hydrodynamic pressure will trigger low-frequency resonance, which will cause the dynamic anti-sway system of a single vessel to be severely disturbed by the underlying environment.

[0137] Based on the understanding of the causal relationship between the aforementioned flow field coupling interference, the system has been extended with a multi-ship collaborative and wake vortex anti-disturbance scheduling strategy. This strategy maps the sensing data of local individuals to the global multi-ship system, and its specific execution logic is detailed as follows: S510. As a preferred approach, the primary basis for multi-ship collaborative computing is to eliminate the spatiotemporal misalignment of data between distributed systems.

[0138] In order to quantify the distortion range caused by upstream operation nodes to the downstream flow field, the multi-ship communication bus acquires the absolute spatial coordinates of the upstream operation cage and the corresponding equivalent flow velocity data.

[0139] Considering the inherent network latency of the surface wireless communication link, the downstream control system extracts the built-in high-precision timestamp of the global navigation satellite system as an alignment reference after receiving the data.

[0140] The system uses the Kalman filter algorithm to extrapolate and predict the state space of upstream state data with historical timestamps, thereby compensating for the spatial coordinate position lag caused by network latency and ensuring the absolute alignment of the relative distance calculation between upstream and downstream.

[0141] After completing multi-source spatiotemporal registration, the system calculates the reduced wake velocity along the direction of the water flow based on the bluff body flow theory. The calculation formula is as follows: ; in, Represents a unified registration timestamp The distance from the upstream cage in the downstream direction is as follows: The wake velocity is reduced at the point; This refers to the aligned equivalent flow rate data defined earlier. The equivalent characteristic width of the upstream gabion cage perpendicular to the water flow direction; The empirical wake attenuation coefficient is determined in the engineering flume based on the porosity characteristics of the gabion. The parameter setting range mentioned above is directly referenced here.

[0142] Considering the distance in the downstream direction when the two cages are extremely close together There is a risk of the value approaching zero. To avoid computational overflow anomalies caused by minimizing the denominator in division operations, a maximum value function is introduced into the denominator of the formula. and the boundary of the minimum value .

[0143] This boundary value is set based on the thickness of the protective shell of the cage body and is always a positive real number.

[0144] Meanwhile, to prevent the reduction ratio calculated at extremely short distances from having a negative value that defies common sense, the system nests a function that compares with zero to take the maximum value outside the ratio term, ensuring that the calculated wake velocity attenuates to the zero dead water zone at the minimum.

[0145] The calculation model deduces the degree of attenuation of the downstream flow field kinetic energy by the upstream permeable component and the disturbance gradient.

[0146] S520. After obtaining the reduced wake velocity in the continuous space, the system needs to further define the isolation boundary for multi-ship collaborative operations.

[0147] The system compares the difference between the reduced wake velocity and the aligned equivalent velocity of the free flow field. To avoid severe boundary jitter caused by hydrological disturbances, the flow field disturbance immunity module is equipped with a decision hysteresis window that includes upper and lower boundaries.

[0148] When the wake reduction velocity varies with distance Only when the increase gradually recovers and remains stably above the set recovery ratio threshold of the aligned equivalent flow velocity for three consecutive calculation cycles, does the system define the corresponding downstream distance as the wake influence limit length. In this embodiment, the recovery ratio threshold is set to 90%.

[0149] Combining the fluid diffusion angle set in fluid mechanics, the properties of which are constrained by the shape of the net cage, in this embodiment, based on the empirical constant of hydrodynamics, the value is usually between 10° and 15°. The system generates a fan-shaped wake vortex anti-disturbance zone with the upstream net cage as the vertex and the wake influence limit length as the center height in the two-dimensional projection plane.

[0150] The area is directly encapsulated as a dynamic electronic fence in the underlying control logic and broadcast synchronously in real time to all cooperating vessels in the same operating waters via high-frequency radio.

[0151] S530. After receiving the dynamic electronic fence data broadcast from the outside, the control system of the downstream crane vessel forces a trajectory interference rehearsal before executing the lowering action.

[0152] The system extracts its own cage coordinate data and the target drop trajectory, and uses a polygonal interferometry detection algorithm to compare the predicted spatial envelope of the cage to be dropped with the broadcast wake vortex anti-disturbance zone.

[0153] For the geometric solution of collision interference of spatial polygons, those skilled in the art can use a boundary interference detection algorithm based on the separating axis theorem.

[0154] S540. When it is determined that the planned descent trajectory of the downstream cage intrudes into the above-mentioned disturbance zone, the control system actively cuts off the current descent execution command and allocates the final avoidance strategy based on multi-dimensional state weighted logic.

[0155] To avoid secondary mechanical damage or serious downtime caused by relying solely on extreme value determination, a comprehensive cost assessment model for quantifying avoidance costs was constructed.

[0156] Specifically, the system calculates the corresponding execution cost by combining the current hovering flow velocity of the water layer, the project schedule margin, and energy consumption parameters: ; in, To avoid sacrificing value for the system as a whole; , , These are the normalized weighting coefficients representing the priorities of safety, efficiency, and energy consumption, respectively. A flexible fatigue risk factor that is positively correlated with hovering flow velocity; This is a project delay penalty factor based on the estimated hovering waiting time; This refers to the fuel consumption factor related to the ship's translational distance.

[0157] If the calculation and comparison results show that the comprehensive cost of peak-shifting hovering is the smallest, and it is determined that the current hovering flow rate has not exceeded the safety tensile limit of the flexible sling of the cage, the system generates a peak-shifting command.

[0158] The execution module then controls the winch to apply the brake, keeping the net cage suspended at the current safe water level until it detects that the upstream net cage has touched the bottom and is in place and the local flow field has stabilized before resuming its descent.

[0159] Conversely, if the weighted assessment shows that the fatigue risk caused by the current hovering flow velocity in the water layer is extremely high, resulting in the time-shifting cost being inferior to the detour cost, the system determines that the shifting scheme is invalid.

[0160] At this point, the system generates a space detour command, directly calling the underlying translation logic of the flow field anti-interference module mentioned above, driving the ship to carry the net cage to move laterally to a safe water area outside the dynamic electronic fence, and replans the release path.

[0161] Through the aforementioned dynamic cost assessment mechanism that is not a single extreme value, the system resolves the hydrodynamic interference caused by dense multi-ship operations at the underlying logic, ensuring the safe convergence of the large-scale group control hoisting process.

[0162] See attached document Figure 7 In this embodiment, after completing the core calculations such as dynamic anti-sway, bottom boundary avoidance, free flow field compensation and multi-ship collaborative scheduling, the control system needs to convert the multi-source control signals with potential overlap into specific execution actions of the equipment.

[0163] Because the lifting system involves mechanical structures with large inertia and heterogeneous actuators, directly issuing unverified independent commands can easily lead to rigid interference between mechanical components or overload of servo drives.

[0164] Based on the aforementioned engineering constraints, the system's embedded execution output module integrates instruction fusion with the underlying hardware mapping logic to complete the closed loop from abstract control algorithms to mechanical actions. The specific execution actions and mapping scheduling process are detailed below: S610. As a preferred instruction management strategy, when the system concurrently receives spatial displacement instructions from different upper-level computing modules, including compensation displacement generated by anti-sway instructions, translation instructions output by the flow field anti-disturbance module, and feedforward displacement instructions in the free flow field mode, the execution output module performs fusion calculations on them based on the principle of linear superposition of spatial vectors.

[0165] The system performs vector summation calculations on the horizontal components of each command in a three-dimensional geodetic coordinate system to generate the final comprehensive target displacement vector. To avoid instantaneous displacement abrupt changes exceeding the mechanical system's bearing capacity due to the superposition of multiple extreme working conditions, a hard constraint limiter based on the system's dynamic boundary is set up inside the module.

[0166] The limiter performs threshold truncation on the comprehensive target displacement vector based on the maximum allowable translation speed and maximum load acceleration parameters specified on the nameplate of the current lifting equipment, ensuring that the output command is strictly constrained within the engineering safety envelope.

[0167] If the fusion computing process detects that control commands of different dimensions have reverse overlap in spatial components or logical mutual exclusion conflicts that exceed degrees of freedom, the system performs hardware arbitration based on the preset mandatory safety priority state machine.

[0168] In this priority allocation setting, multi-ship collaborative collision avoidance and slot avoidance involving collisions are the highest level, dynamic anti-sway is the second highest level, and free flow field feedforward compensation is the lowest level.

[0169] The system ensures the unique issuance of core security instructions by blocking control signals of low-priority channels at the software level, thereby preventing the potential for mechanical tearing at the logical source.

[0170] S620. The integrated target displacement vector after arbitration fusion and amplitude limiting constraint needs to be mapped to a specific actuator.

[0171] The main positioning mechanism responsible for macroscopic position movement includes the boom's luffing hydraulic cylinder and slewing motor.

[0172] The execution output module calls the crane's forward and inverse kinematics model, and inversely calculates the comprehensive target displacement vector in the rectangular coordinate system into the expected stroke extension of the luffing hydraulic cylinder and the expected rotation angle of the slewing motor.

[0173] Considering that when the lifting boom is close to its maximum or minimum working angle, its kinematic Jacobian matrix is ​​prone to fall into a singular state due to the determinant approaching zero, which in turn leads to divergence in the inverse solution.

[0174] To address the incompleteness of the algorithm's logic, the system introduces a damped least squares method to regularize the Jacobian matrix during inverse kinematics solving. The damping coefficient is used to limit the joint velocity output near singular points, ensuring that the solution converges absolutely within the entire working envelope.

[0175] The system uses the expected value calculated above as a given input signal and transmits it to the underlying electro-hydraulic servo controller. Addressing the typical nonlinearity and pressure time-delay characteristics of large hydraulic systems, the electro-hydraulic servo controller internally relies on a proportional-integral-derivative closed-loop control algorithm with a feedforward channel to calculate and generate a pulse-width modulated electrical signal for controlling the valve core opening of the proportional servo valve. This signal then drives the crane boom to produce corresponding mechanical translation to accurately execute the upper-level control commands.

[0176] For the specific parameter tuning method of the electro-hydraulic servo control circuit and the working process of the pulse width modulation signal generator, those skilled in the art can use conventional frequency domain analysis or on-site experimental trial methods to set them.

[0177] S630: Independently issues attitude yaw commands to the flow field disturbance suppression module to change the angle of the underwater gabion cage's water-facing surface. These commands are assigned to independent execution branches. The actuator is the attitude adjustment mechanism embedded in the spreader system, specifically manifested as servo hydraulic push rods arranged symmetrically around the universal joint of the spreader and a rotary drive motor integrated on the top of the spreader.

[0178] The execution output module calculates the precise stroke position required by each hydraulic push rod based on the target compensation attitude angle data contained in the attitude yaw command, combined with the coordinates of the mechanical hinge point of the universal joint and the link length parameter, relying on the spatial geometric projection transformation relationship.

[0179] The system independently drives the rotary drive motor to generate yaw torque by sending control electrical signals. At the same time, it combines the differential telescopic displacement generated by the diagonally arranged hydraulic push rods to maintain horizontal balance. This creates a deflection torque at the connection point of the lifting device that can overcome the local fluid torque, forcing the suspended underwater cage to deflect in a rigid body attitude at the expected angle.

[0180] The attitude adjustment mechanism is completely decoupled from the aforementioned main positioning mechanism in terms of spatial layout and power hydraulic circuit, ensuring that the local attitude fine adjustment of the cage and the macro position compensation action of the boom can be carried out simultaneously and without interference.

[0181] After the S640 and underlying execution actions are issued and driven, the status feedback of the actuator constitutes the final link in the system control cycle.

[0182] The system utilizes high-precision magnetostrictive displacement sensors installed on each variable-amplitude hydraulic cylinder and attitude servo push rod, as well as an absolute encoder at the end of the rotary motor shaft, to collect the current actual stroke or angular displacement of various actuators in real time.

[0183] The actual displacement, along with the real-time hydrological and attitude data acquired by the preceding state sensing module, is synchronously transmitted back to the main control unit via the fieldbus, forming a closed-loop monitoring network for the global hardware status.

[0184] If the monitoring network detects that the deviation between the actual response trajectory of the actuator and the expected trajectory of the command exceeds the set dynamic tolerance margin, in this embodiment, considering the inherent mechanical dead zone and hydraulic compression hysteresis characteristics of the hydraulic system, this dynamic tolerance margin is typically set within the range of 5cm to 15cm. Furthermore, if the duration of this overshoot state exceeds the set fault assessment period, which in this embodiment is set to 200ms to 500ms based on the bus communication sampling rate, the system determines that the underlying hardware has experienced jamming or that the main hydraulic power source has failed.

[0185] In such extreme failure conditions, the control system directly triggers the emergency stop protection mechanism through the interrupt pin at the bottom layer of the main control board, immediately locking the hydraulic balance valve and brake mechanism in all current action directions to maintain pressure and lock them in a locked state, and pops up the fault location and audible and visual alarms in real time through the human-machine interface.

[0186] By establishing a complete closed-loop information link from top-level anti-interference algorithm to actuation mapping and then to hardware status monitoring, the system reliably executes multi-dimensional control strategies at the underlying engineering application level, ensuring the execution and positioning of cage hoisting operations in deep-water complex flow field environments.

Claims

1. A lifting and positioning control system for steel gabion cages on water, characterized in that, include: The status awareness module is used to acquire real-time data on water flow, cage depth, sling length, cage coordinates, and lifting gear attitude in the hoisting environment. The dynamic anti-sway module is used to calculate dynamic damping parameters based on the sling length data and the water flow data, and output anti-sway commands based on the dynamic damping parameters; The flow field anti-interference module is used to compare the coordinate data of the cage with the preset three-dimensional construction grid coordinate library. When it is determined that the distance between the current cage and the cage that has been placed is greater than or equal to a set threshold, a feedforward displacement command is generated. When it is determined that the distance between the current net cage and the net cage that has been placed is less than a set threshold, the local flow field disturbance force on the net cage is evaluated based on the water flow data and the attitude data, and the hydrodynamic compensation amount required to offset the local flow field disturbance force is calculated based on the local flow field disturbance force. The hydrodynamic compensation amount is converted into attitude yaw command or translation command and output. The execution output module is used to merge the anti-sway command with the displacement command used to control the horizontal movement of the lifting equipment and output it to the main positioning mechanism of the lifting equipment. The displacement command includes the translation command output by the flow field anti-disturbance module. The execution output module is also used to output the attitude yaw command to the attitude adjustment mechanism of the lifting device.

2. The waterborne steel gabion cage hoisting and positioning control system according to claim 1, characterized in that, The state perception module acquires vertical water flow velocity profile data as the water flow data through underwater acoustic flow measurement equipment, acquires the sling length data through the encoder of the hoisting winch, acquires the ship's reference coordinates through the global satellite navigation system receiver, and calculates the net cage coordinate data by combining the boom kinematic parameters. It also acquires the net cage depth data and attitude data through the depth gauge and inertial measurement unit installed on the lifting device. The attitude data includes pitch angle, roll angle, angular velocity vector, and three-axis acceleration vector.

3. The waterborne steel gabion cage hoisting and positioning control system according to claim 2, characterized in that, The dynamic anti-sway module is specifically used for: Based on the depth data of the gabion cage, the corresponding aligned equivalent flow velocity data is extracted from the water flow data, and the fluid added mass is calculated by combining the preset porosity parameters of the gabion cage. The fluid-added mass is used as a mass parameter to update the system's preset mass matrix, while the real-time updated cable length data is used as a scheduling variable to construct a linear parameter change state space model. Based on the linear parameter change state space model and the real-time angular velocity vector extracted from the attitude data, the regulator gain matrix is ​​solved to generate the anti-sway command used to suppress the flexible sway of the hoisting equipment.

4. The waterborne steel gabion cage hoisting and positioning control system according to claim 2, characterized in that, The local flow field disturbance force is the local adsorption force in the slit caused by adjacent cages; The flow field anti-disturbance module has a built-in inverse dynamics disturbance observer. When it is determined that the distance between the current cage and the cage that has been placed is less than the set threshold, the inverse dynamics disturbance observer uses the components of the triaxial acceleration vector in the horizontal dimension of the attitude data, the preset total system mass and the acquired control driving force to calculate the local adsorption force of the slit that represents the real environmental disturbance in real time through the inverse dynamics equation.

5. The waterborne steel gabion cage hoisting and positioning control system according to claim 4, characterized in that, When the flow field disturbance rejection module outputs the attitude yaw command or the translation command, it specifically executes the following dynamic allocation logic: The maximum lateral hydrodynamic lift that the cage can generate is calculated based on the aligned equivalent flow velocity data extracted from the water flow data based on the cage depth data and the maximum yaw angle allowed by the attitude adjustment mechanism. When the maximum lateral hydrodynamic lift is greater than or equal to the local suction force of the slit, the system only generates the attitude yaw command, and uses the lateral hydrodynamic lift generated by the tilting of the cage to face the water to counteract the local suction force of the slit. When the maximum lateral hydrodynamic lift is less than the local suction force of the slit, the system locks the attitude yaw command to the control signal corresponding to the maximum yaw angle, and converts the difference between the local suction force of the slit and the maximum lateral hydrodynamic lift into the translation command.

6. The waterborne steel gabion cage hoisting and positioning control system according to claim 5, characterized in that, The flow field disturbance rejection module is equipped with a zero-vibration input shaper. When the system outputs the translation command, the zero-vibration input shaper calculates the real-time natural frequency based on the sling length data and the water buoyancy correction coefficient, and decomposes the step-form translation command into multiple pulse displacement sequences with preset time delays, so that the primary oscillation and secondary oscillation caused by the translation of the main positioning mechanism cancel each other out.

7. The waterborne steel gabion cage hoisting and positioning control system according to claim 5, characterized in that, When the maximum lateral hydrodynamic lift is greater than or equal to the local adsorption force of the slit, the flow field anti-disturbance module, based on the fluid lift inverse solution model, uses the aligned equivalent flow velocity data, the pre-calibrated attitude angle and lift coefficient mapping relationship, and the local adsorption force of the slit to be offset to inversely calculate the target compensation attitude angle, and outputs the target compensation attitude angle as the attitude yaw command.

8. The waterborne steel gabion cage hoisting and positioning control system according to claim 1, characterized in that, When it is determined that the distance between the current cage and the cage that has been placed is greater than or equal to the set threshold, the flow field anti-disturbance module stops outputting the attitude yaw command and the translation command; The system extracts the corresponding aligned equivalent flow velocity data from the water flow data based on the cage depth data, calculates the macroscopic water flow thrust by combining the orthographic projection area and porosity parameters of the cage, and generates a feedforward displacement command as the displacement command. The execution output module merges the feedforward displacement command with the anti-sway command and outputs it to the main positioning mechanism.

9. A waterborne steel gabion hoisting and positioning control system according to claim 1, characterized in that, The system also includes a network scheduling module, which is connected to the status sensing module and is used to receive the operating coordinates and lowering depth of adjacent lifting vessels in the same construction water area. When the upstream and downstream crane vessels in the same construction water area are identified based on the coordinate data of the net cage and the operation coordinates, and it is calculated that the downstream crane vessel is within the diffusion envelope of the wake vortex generated by the upstream crane vessel lowering the net cage, a deep hovering command is issued to the winch control system of the downstream crane vessel until the local flow field is determined to have returned to steady state and the hovering is released.

10. A waterborne steel gabion hoisting and positioning control system according to claim 1, characterized in that, The main positioning mechanism is the slewing mechanism and luffing mechanism of the hoisting vessel, and the attitude adjustment mechanism is a hydraulic hoisting frame equipped with four independent hydraulic leveling cylinders at the corners and a slewing drive motor. The execution output module converts the fused command into a pulse width modulation electrical signal and sends it to the slewing mechanism and the amplitude conversion mechanism. It also converts the attitude yaw command into a control electrical signal and sends it to the slewing drive motor. The four independent hydraulic leveling cylinders at the corners are used to maintain the horizontal attitude.

Images